Ex Parte Nanavati et al

Patent Trial and Appeal Board

Jun 6, 2018

13312511 (P.T.A.B. Jun. 6, 2018)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
13/312,511 12/06/2011
89885 7590 06/06/2018
FERENCE & ASSOCIATES LLC
409 BROAD STREET
PITTSBURGH, PA 15143
FIRST NAMED INVENTOR
Amit A. Nanavati
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www .uspto.gov
ATTORNEY DOCKET NO. CONFIRMATION NO.
IN9201 l 0205US 1
(790.141)
1097
EXAMINER
ALVAREZ, RAQUEL
ART UNIT PAPER NUMBER
3688
MAILDATE DELIVERY MODE
06/06/2018 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte AMIT A. NANAVATI and RAMASURI NARAYANAM
Appeal2017-001814 1
Application 13/312,511
Technology Center 3600
Before ANTON W. PETTING, NINA L. MEDLOCK, and
AMEE A. SHAH, Administrative Patent Judges.
SHAH, Administrative Patent Judge.
DECISION ON APPEAL 2
The Appellants3 appeal under 35 U.S.C. § 134(a) from the Examiner's
final decision rejecting claims 13-17, 19, 22, and 25-29, which are all of the
pending claims. We have jurisdiction under 35 U.S.C. § 6(b ).
We AFFIRM.
1 We note related appeal 2017-001831, application 13/599,018.
2 Throughout this Decision, we refer to the Appellants' Appeal Brief
("Appeal Br.," filed Apr. 5, 2016), Reply Brief ("Reply Br.," filed Nov. 15,
2016), and Specification ("Spec.," filed Dec. 6, 2011), and to the Examiner's
Answer ("Ans.," mailed Sept. 15, 2016), and Final Office Action ("Final
Act.," mailed Nov. 5, 2015).
3 According to the Appellants, the real party in interest is International
Business Machines Corporation. Appeal Br. 3.
Appeal2017-001814
Application 13/312,511
STATEMENT OF THE CASE
The Appellants' invention relates to "methods and arrangements for
facilitating viral marketing campaigns that lend themselves to significant
cross-selling and/ or up-selling." Spec. i-f 17.
Claims 13 and 14 are the independent claims on appeal. Claim 13
(Appeal Br. 30-31 (Claims App.)) is exemplary of the subject matter on
appeal and is reproduced below (lettered bracketing added for reference).
13. An apparatus comprising:
at least one processor; and
a non-transitory computer readable storage medium
having computer readable program code embodied therewith and
executable by the at least one processor, the computer readable
program code comprising:
[(a)] computer readable program code configured to
ascertain participants in a social network;
[(b )] computer readable program code configured to
determine mutual influence of the participants in purchasing
decisions;
[ ( c)] computer readable program code configured to assess
a prospective influence of at least a first product on purchasing
at least a second product; and
[ ( d)] computer readable program code configured to select
at least one seed from the participants for maximizing influence
of the first product on purchasing the second product, the
selecting being based on:
the assessed prospective influence of the first
product on purchasing the second product; and
the determined mutual influence of at least one
participant on at least one other participant;
2
Appeal2017-001814
Application 13/312,511
[ ( e)] the selecting compnsmg employing a selling
approximation algorithm, the selling approximation algorithm
comprising one or more of:
a maximum degree based heuristic algorithm; and
maximum influence based heuristic algorithm.
REJECTIONS
I. Claims 13-17, 19, 22, and 25-29 stand rejected under
35 U.S.C. § 101 as being directed to non-statutory subject matter.
II. Claims 13-17, 19, 22, and 25-294 stand rejected under the
doctrine of provisional nonstatutory double patenting over claims 1--4, 6, 9,
and 13-16 of co-pending application number 13/599,018, in view of
Levy et al. (US 2011/0282821 Al, pub. Nov. 17, 2011) (hereafter "Levy").
III. Claims 13-17, 19, 22, and 25-29 stand rejected under
35 U.S.C. § 103(a) as being unpatentable over Paul et al.
(US 2009/0063254 Al, pub. Mar. 5, 2009) (hereafter "Paul"), Walker et al.
(US 2009/0198625 Al, pub. Aug. 6, 2009) (hereafter "Walker"), and Levy.
ANALYSIS
Rejection I-Patent-Ineligible Subject Matter-§ 101
The Appellants argue claims 13-17, 19, 22, and 25-29 as a group.
See Appeal Br. 15. We select claim 13 as representative of the group;
claims 14--17, 19, 22, and 25-29 stand or fall with claim 13. See 37 C.F.R.
§ 41.37(c)(l)(iv).
4 We consider the Examiner's omission of claim 13 (Final Act. 6) to be
inadvertent error (see Ans. 2 (using claim 13 in the comparison)).
3
Appeal2017-001814
Application 13/312,511
Under 35 U.S.C. § 101, a patent may be obtained for "any new and
useful process, machine, manufacture, or composition of matter, or any new
and useful improvement thereof." The Supreme Court has "long held that
this provision contains an important implicit exception: Laws of nature,
natural phenomena, and abstract ideas are not patentable." Alice Corp. Pty.
Ltd. v. CLS Bank Int 'l, 134 S. Ct. 2347, 2354 (2014) (quoting Ass 'nfor
Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576, 588-89
(2013)).
The Supreme Court in Alice reiterated the two-step framework, set
forth previously in Mayo Collaborative Services v. Prometheus
Laboratories, Inc., 566 U.S. 66, 78-79 (2012), "for distinguishing patents
that claim laws of nature, natural phenomena, and abstract ideas from those
that claim patent-eligible applications of those concepts." Alice, 134 S. Ct.
at 2355. The first step in that analysis is to "determine whether the claims at
issue are directed to one of those patent-ineligible concepts." Id. (emphasis
added) (citing Mayo, 566 U.S. at 79). If so, the second step is to consider
the elements of the claims "individually and 'as an ordered combination"' to
determine whether the additional elements "'transform the nature of the
claim' into a patent-eligible application." Id. (quoting Mayo, 566 U.S.
at 78-79).
We first address the Appellants' argument that the Examiner "has
failed to articulate how the claims are directed to an abstract idea and fail to
amount to significantly more than an abstract idea, but rather has made an
unsupported assertion of such" (Appeal Br. 15; see also Reply Br. 14), and
has thus "failed to present a prima facie case of unpatentability" (Appeal
Br. 16; Reply Br. 15). Specifically, the Appellants argue that the Examiner
4
Appeal2017-001814
Application 13/312,511
"fails to specifically point out the factors that are relied upon in making the
determination that the claims are directed to an abstract idea," "has not
identified the specific limitations considered or any rationale as to why these
limitations are not enough to qualify as 'significantly more,"' and "has
ignored other claim limitations and relevant factors in reaching such []
determination[s]." Appeal Br. 16; Reply Br. 14--15. We disagree.
Here, in rejecting claim 13 under § 101, the Examiner analyzes the
claim using the Mayo/Alice two-step framework. Specifically, the Examiner
looks to the intrinsic evidence of the claim language as factors in
determining that the claim is directed to an abstract idea. See Ans. 3; Final
Act. 2. The Examiner also cites to judicial decisions and compares the
concept to those found to be abstract in those decisions. Ans. 3--4. The
Examiner further considers the claim's limitations individually and as an
ordered combination as factors in determining that the claim does not recite
limitations that transformed the nature of the claim into a patent-eligible
invention. See Ans. 4--5; Final Act. 3. The Examiner cites to and compares
the elements to judicial decisions as support and evidence for determining
that the claim does not contain an inventive concept. See Ans. 4--5.
Thus, the Examiner has clearly articulated the reasons as to why the
claim is directed to an abstract idea and has notified the Appellants of the
reasons for the rejection "together with such information and references as
may be useful in judging of the propriety of continuing the prosecution of
[the] application." 35 U.S.C. § 132. In doing so, the Examiner has set forth
a prima facie case ofunpatentability. See In re Jung, 637 F.3d 1356, 1362
(Fed. Cir. 2011); Chester v. Miller, 906 F.2d 1574, 1578 (Fed. Cir. 1990)
(Section 132 "is violated when a rejection is so uninformative that it
5
Appeal2017-001814
Application 13/312,511
prevents the applicant from recognizing and seeking to counter the grounds
for rejection"). We further note that the Examiner has clearly followed the
Office's guidelines.
Step One of the Mayo/Alice Framework
Turning to the first step of the Mayol Alice framework, we consider the
claims "in their entirety to ascertain whether their character as a whole is
directed to excluded subject matter." Internet Patents Corp. v. Active
Network, Inc., 790 F.3d 1343, 1346 (Fed. Cir. 2015). The question is
whether the claims as a whole "focus on a specific means or method that
improves the relevant technology" or are "directed to a result or effect that
itself is the abstract idea and merely invoke generic processes and
machinery." McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299,
1314 (Fed. Cir. 2016). In other words, the claims are assessed as to whether
they "do no more than describe a desired function or outcome, without
providing any limiting detail that confines the claim to a particular solution
to an identified problem." Affinity Labs of Tex., LLC v. Amazon.com Inc.,
838 F.3d 1266, 1269 (Fed. Cir. 2016).
Here, the Examiner determines that claim 13 is directed to "the
abstract idea of: determining influence of participants in purchasing
decisions." Ans. 3; see also Final Act. 2. Conversely, the Appellants
contend the claim is directed to "automated processing techniques for
selecting a participant in a social network to act as a seed for maximizing
influence of purchasing a product while taking into account a selling
approximation algorithm and a mutual influence." Appeal Br. 17.
The Title of the Specification provides for "DESIGNING VIRAL
MARKETING STRATEGIES FOR UP-SELLING AND CROSS-
6
Appeal2017-001814
Application 13/312,511
SELLING." The Background section of the Specification discusses that
"[g]enerally, viral marketing has emerged as an effective tool for marketing
in view of the increasing popularity of online social networks." Spec. i-f 1.
The Background section also discusses that limited advertising budgets
present a "key challenge" of selecting a set of influential individuals/seeds in
the network and putting them in a position to raise awareness of a product
over that network. Id. The claimed invention addresses this problem by
providing for "methods and arrangements for facilitating viral marketing
campaigns that lend themselves to significant cross-selling and/or up-
selling." Id. i-f 17. Figure 4, to which the Appellants refer as support for
claim 13 (Appeal Br. 7-8), provides for "a process more generally for
instituting a marketing strategy for cross-sell or upsell." Spec. i-f 30; see also
id. ,-r 9.
Claim 13 provides for an apparatus comprising a processor and a
medium comprising code configured to perform the functions of:
(a) "ascertain[ing] participants in a social network," (b) "determin[ing]
mutual influence of the participants in purchasing decisions,"
( c) "assess[ing] a prospective influence" of a first product on a second
products, and ( d) selecting a seed from the participants based on the
determines mutual influence and the assessed prospective influence, ( e) the
selecting comprising employing a heuristic algorithm. Appeal Br. 30-31
(Claims App.). The Appellants refer generally to paragraphs 18, 19, 21, 27,
28, 30, and 31 for providing support for the claim as a whole, but do not map
the limitations to the specific paragraphs in Specification that provide details
for each of the limitations. See Appeal Br. 7-12. But we do not see where
the Specification specifically describes how, i.e., in what way, the
7
Appeal2017-001814
Application 13/312,511
participants are ascertained, how the mutual influence is determined, or how
the purchasing influence is assessed. See Spec. i-f 31. Further, the
Specification provides that all of the claimed functions "can be carried out
on essentially any suitable computer system or set of computer systems." Id.
i-f 30; see also id. i-fi-1 16, 33--40 (discussing a cloud computing node
operational with general purpose, well-known computing systems having
conventional components).
In light of Specification's description of the problem and solution, the
purported advance over the prior art by the claimed invention is a way to
present better marketing campaigns that lend themselves to upselling and/or
cross-selling of products based on social networks and influence data. In
that context, claim 13 is directed to marketing to upsell or cross-sell products
based on the analysis of influence data. 5 The claim here is akin to ones our
reviewing court deemed abstract in Intellectual Ventures I LLC v. Capital
One Bank (USA), 792 F.3d 1363, 1369--70 (Fed. Cir. 2015) (customizing
and tailoring web page content based on navigation history and known user
information), Electric Power Group LLC v. Alstom S.A., 830 F.3d 1350,
1353-54 (Fed. Cir. 2016) (collecting information and "analyzing
information by steps people go through in their minds, or by mathematical
algorithms, without more, ... [are] essentially mental processes within the
abstract-idea category"), and Affinity Labs., 838 F.3d at 1271 (customizing a
user interface to have targeted advertising based on user information). Here,
5 We note that "[a Jn abstract idea can generally be described at different
levels of abstraction." Apple, Inc. v. Ameranth, Inc., 842 F.3d 1229, 1240
(Fed. Cir. 2016). The Board's "slight revision of its abstract idea analysis
does not impact the patentability analysis." Id. at 1241.
8
Appeal2017-001814
Application 13/312,511
the claim involves nothing more than a generic computer to determine and
analyze data, and select data based on the analysis, without any particular
inventive technology- an abstract idea. See Elec. Power, 830 F.3d at 1354
see also Alice, 134 S. Ct. at 2360 ("none of the hardware recited by the
system claims 'offers a meaningful limitation beyond generally linking 'the
use of the [method] to a particular technological environment,' that is,
implementation via computers."') (alteration in original).
Further, marketing to upsell and cross-sell products is a fundamental
building block of modem economy. See, e.g., Wibke Heidig, Upselling or
Upsetting? Studies on the Behavioral Consequences of Upsell Offers in
Service Encounters (Oct. 29, 2012) (unpublished Ph.D dissertation,
University of St. Gallen) (available at
https://wwwl.unisg.ch/www/edis.nsf/SysLkpByidentifier/4086/$FILE/dis40
86.pdf); Larry Levine, Why Cross-Selling and Upselling Seem So Difficult to
Implement, Telemarketing & Call Center Solutions, Jan. 1, 1996 (HighBeam
Research) (hereafter "Levine"). As such, we find unpersuasive the
Appellants' arguments that the claim is not directed to an abstract idea.
Appeal Br. 18; Reply Br. 16-20, 23.
In response to the Appellants' argument that "there is clearly no risk
that the claims will 'tie up' the subject matter of selecting a participant in a
social network or pre-empt others from employing alternate methods of
selecting a participant in a social network" (Appeal Br. 18; Reply Br. 19),
we note that although the Supreme Court has described "the concern that
drives this exclusionary principle [i.e., the exclusion of abstract ideas from
patent eligible subject matter] as one of preemption" (see Alice, 134 S. Ct.
at 2354), characterizing preemption as a driving concern for patent eligibility
9
Appeal2017-001814
Application 13/312,511
is not the same as characterizing preemption as the sole test for patent
eligibility. "The Supreme Court has made clear that the principle of
preemption is the basis for the judicial exceptions to patentability" and "[ fJor
this reason, questions on preemption are inherent in and resolved by the
§ 101 analysis." Ariosa Diagnostics, Inc. v. Sequenom, Inc., 788 F.3d 1371,
1379 (Fed. Cir. 2015) (citing Alice, 134 S. Ct. at 2354). Although
"preemption may signal patent ineligible subject matter, the absence of
complete preemption does not demonstrate patent eligibility." Id. The
claimed invention is not sufficiently limiting so as to fall clearly on the side
of patent-eligibility.
We also find unpersuasive the Appellants' arguments that the claim is
not directed to an abstract idea because it is analogous to those claims of
DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014).
Appeal Br. 19--20; Reply Br. 20-21. In DDR Holdings, the Federal Circuit
determined that the claims addressed the problem of retaining website
visitors who, if adhering to the routine, conventional functioning of Internet
hyperlink protocol, would be transported instantly away from a host's
website after clicking on an advertisement and activating a hyperlink. DDR
Holdings, 773 F.3d at 1257. The Federal Circuit, thus, held that the claims
were directed to statutory subject matter because they claim a solution
"necessarily rooted in computer technology in order to overcome a problem
specifically arising in the realm of computer networks." Id. The court
cautioned that "not all claims purporting to address Internet-centric
challenges are eligible for patent." Id. at 1258. And the court contrasted the
claims to those at issue in Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709
(Fed. Cir. 2014), in that, in DDR Holdings, the computer network was not
10
Appeal2017-001814
Application 13/312,511
operating in its "normal, expected manner" and the claims did not "recite an
invention that is ... merely the routine or conventional use of the Internet."
Id. at 1258-59.
In contrast, here, the Appellants do not specify what problem
"specifically arising in the realm of computer networks" the claim addresses.
As noted above, the Specification discusses existing problems of limited
advertising budgets that present a "key challenge" of selecting a set of
influential individuals/seeds in the network and putting them in a position to
raise awareness of a product over that network. Spec. i-f 1. Although the
social networks are network-centric, selecting influential individuals and
raising awareness of a product are not problems rooted in technology or
specifically arising in computer networks, but rather business problems that
existed prior to the Internet and computers. See Levine (discussing issues
with selling, upselling, and cross-selling when customers call using
telephones).
Also, unlike DDR Holdings, here, the purported solution to address
this problem is an "arrangement[] for facilitating viral marketing campaigns
that lend themselves to significant cross-selling and/or up-selling" (Spec.
i-f 17) that comprises a generic processor operating in its normal capacity to
execute code that achieves the desired business-based result of maximizing
influence of a product by determining and analyzing data. See Spec. i-fi-1 16,
30, 41--48, Fig. 5. The Appellants do not direct attention to, and we do not
see, where the Specification describes the processor or other computer
components acting in an unconventional manner to further the desired
solution of cost-effective advertising. Rather, the claims "recite an invention
11
Appeal2017-001814
Application 13/312,511
that is ... merely the routine or conventional use of the Internet." DDR
Holdings, 773 F.3d at 1258-59.
We further find unpersuasive the Appellants' argument that the
Examiner's "attempted reliance on the fact that the claims might be
implemented on a 'generic computer' to find an abstract idea is misplaced"
(Reply Br. 22), because the claim "at the very least represent[s] a software
improvement to social network participant selection that is novel and
nonobvious as compared to prior art techniques" (id. at 23 (citing Enfish
and McRO)). 6 The claims at issue in Enfish were directed to a specific type
of data structure, i.e., a self-referential table for a computer database,
designed to improve the way a computer carries out its basic functions of
storing and retrieving data. Enfzsh, LLC v. Microsoft Corp., 822 F.3d 1327,
1335-36 (Fed. Cir. 2016). There, in rejecting a§ 101 challenge, the court
held that "the plain focus of the claims is on an improvement to computer
functionality itself, not on economic or other tasks for which a computer is
used in its ordinary capacity." Id. at 1336. The Appellants have not
adequately explained here how the court's holding in Enfzsh impacts the
present analysis under the Mayo/Alice framework. For example, the
Appellants do not point to anything in the claims that resembles the
inventive self-referential data structure at issue in Enfzsh. And the
Appellants do not direct our attention to anything in the Specification that
indicates that the invention provides an improvement in computer
functionality.
6 The Appellants may present this new argument based on the recent
relevant decision of the Federal Circuit issued after the Appeal Brief.
12
Appeal2017-001814
Application 13/312,511
Moreover, an abstract idea does not transform into an inventive
concept just because the prior art does not disclose or suggest it. See Mayo,
566 U.S. at 91. "Groundbreaking, innovative, or even brilliant discovery
does not by itself satisfy the § 101 inquiry." Association for Molecular
Pathology v. Myriad Genetics, Inc., 133 S. Ct. 2107, 2117 (Fed. Cir. 2013).
Indeed, "[t]he 'novelty' of any element or steps in a process, or even of the
process itself, is of no relevance in determining whether the subject matter of
a claim falls within the § 101 categories of possibly patentable subject
matter." Diamond v. Diehr, 450 U.S. 17 5, 188-89 (1981 ); see also Mayo,
566 U.S. at 91 (rejecting "the Government's invitation to substitute§§ 102,
103, and 112 inquiries for the better established inquiry under§ 101").
Step Two of the Mayol Alice Framework
Turning to the second step of the Mayo/Alice framework, we "search
for an 'inventive concept'-i.e., an element or combination of elements that
is 'sufficient to ensure that the patent in practice amounts to significantly
more than a patent upon the [ineligible concept] itself."' Id. (alteration in
original) (quoting Mayo, 566 U.S. at 72-73). The Court acknowledged in
Mayo, that "all inventions at some level embody, use, reflect, rest upon, or
apply laws of nature, natural phenomena, or abstract ideas." Mayo, 566 U.S.
at 71. We, therefore, look to whether the claims focus on a specific means
or method that improves the relevant technology or are instead directed to a
result or effect that itself is the abstract idea, and merely invoke generic
processes and machinery, i.e., "whether the focus of the claims is on [a]
specific asserted improvement in computer capabilities ... or, instead, on a
process that qualifies as an 'abstract idea' for which computers are invoked
merely as a tool." See Enfzsh, 822 F.3d at 1335-36.
13
Appeal2017-001814
Application 13/312,511
We agree with and adopt the Examiner's determination that the
elements of claim 13, individually or as an ordered combination, do not
amount to significantly more than the above-identified abstract idea. See
Final Act. 3; Ans. 3-5. We note, as discussed above, that the Specification
conveys that the computer-related component recited in the claim (e.g.,
"processor") can be "essentially any suitable computer system or set of
computer systems, which may, by way of an illustrative and non-restrictive
example, include a system or server such as that indicated at 12' in FIG. 5"
(Spec. i-f 16), that "computer system/server 12' in cloud computing node 10 is
shown in the form of a general-purpose computing device" including a
processor (id. i-f 35), and the "computer program instructions may be
provided to a processor of a general purpose computer" (id. i-f 46).
The Appellants argue that the claim "add[s] 'significantly more' to
selecting a participant to maximize purchasing influence by adding a
capability that the participant is selected based upon a selling approximation
algorithm and a mutual influence. Claim 13. This is in contrast to
conventional techniques applied in this field." Appeal Br. 21; Reply Br. 24.
To the extent the Appellants argue that the claim is significantly more
than the abstract idea because it is not taught by the prior art, we emphasize
that our review for patent-eligible subject matter under 35 U.S.C. § 101 is
independent of any review of another rejection. As discussed above, "[t]he
'novelty' of any element or steps in a process, or even of the process itself, is
of no relevance in determining whether the subject matter of a claim falls
within the§ 101 categories of possibly patentable subject matter."
Diamond, 450 U.S. 188-89.
14
Appeal2017-001814
Application 13/312,511
To the extent the Appellants argue that selecting based upon an
algorithm and influences is not conventional, we disagree. Making
selections using an algorithm and based on data is a routine, well-
understood, and conventional activity of a generic computer that does not
transform the claimed subject matter into patent-eligible applications. See
Elec. Power, 830 F.3d at 1354--55 (gathering, sending, monitoring,
analyzing, selecting, and presenting information does not transform the
abstract process into a patent-eligible invention); OIP Technologies, Inc. v.
Amazon.com, Inc., 788 F.3d 1359, 1363---64 (Fed. Cir. 2015). The
Appellants do not provide further support or reasoning as to why or how the
limitations of selecting a seed based on an algorithm and influence data are
not well-understood, routine, and conventional functions of a generic
computer. We note that computer technology itself is not improved. Any
improvement resides in the routine tasks of selecting data.
Based on the foregoing, we sustain the Examiner's rejection under
35 U.S.C. § 101 of claim 13, and of claims 14--17, 19, 22, and 25-29, which
fall with claim 13.
Rejection II- Provisional Nonstatutory Double Patenting
As our reviewing court had held:
The obviousness-type double patenting analysis involves
two steps: "First, the court 'construes the claim[s] in the earlier
patent and the claim [ s] in the later patent and determines the
differences.' Second, the court 'determines whether those
differences render the claims patentably distinct."' Sun Pharm.
Indus., Ltd. v. Eli Lilly & Co., 611 F.3d 1381, 1385
(Fed.Cir.2010) (alteration in original) (quoting Pfizer, Inc. v.
Teva Pharm. USA, Inc., 518 F.3d 1353, 1363 (Fed. Cir.2008)).
"'A later claim that is not patentably distinct from,' i.e., 'is
15
Appeal2017-001814
Application 13/312,511
obvious over[ ] or anticipated by,' an earlier claim is invalid for
obviousness-type double patenting." Id. at 1385 (alteration in
original) (quoting Eli Lilly [&Co. v. Barr Labs., Inc., 251 F.3d
955,] 968 [(Fed. Cir. 2001)].
Abbvie Inc. v. Mathilda & Terence Kennedy Inst. of Rheumatology Tr.,
764 F.3d 1366, 1374 (Fed. Cir. 2014)
The Examiner finds that the differences between the claims here and
those of co-pending application 13/599,018 ("'018 application") are that
"the instant claims are claiming an apparatus and the copending application
13/599,018 [claims] are claiming method steps" (Ans. 2) and "[t]he instant
claims do not recite a two dimensional determination of influence of a first
product on a second product and the influence the participants have on other
participants" (Final Act. 6). The Examiner then cites to Levy for teaching
"users being used as influencers for cross-selling and influencing others to
buy a second product," and determines that it would have been obvious to
one of ordinary skill in the art at the time of the invention "to have excluded
a two dimensional determination of influence of a first product on a second
product and the influence the participants have on other participants, because
such a modification would will [sic] provide more flexibility." Id.
The Appellants contend that the Examiner's rejection is in error
because "Levy is readily distinguishable from the present application and
claims" and because "the claims fall under two statutorily distinct classes."
Appeal Br. 14; Reply Br. 13-14.
Although the Examiner determines the differences between the two
claim sets, the Examiner does not construe the claims and determine whether
those differences render the claims patentably distinct. See Sun Pharm.
Indus., 611 F.3d at 1385. Further, the Examiner's findings do not
16
Appeal2017-001814
Application 13/312,511
sufficiently address whether the Levy's selecting is based on a non-two-
dimensional analysis or how excluding one provides more flexibility.
For these reasons, we do not sustain the Examiner's rejection of
claims 13-17, 19, 22, and 25-29 under the doctrine of double patenting.
Rejection III - Obviousness-§ 103 (a)
We agree with the Appellants' contention that the Examiner does not
adequately show that the prior art teaches assessing a prospective influence
of a first product on purchasing a second product and selecting a seed based
in part on that assessed prospective influence, as recited in limitations ( c)
and ( d) of independent claim 13, and similarly recited in independent
claim 14. See Appeal Br. 24--27; Reply Br. 27-30.
Regarding the above limitations, the Examiner finds that Paul teaches
selecting a seed "for maximizing of potential of a product purchase by the at
least one seed." Final Act. 8 (citing Paul i-fi-f 12, 39). The Examiner
acknowledges that "Paul is silent as to maximizing upsell/cross selling of
product purchase." Id. The Examiner takes Official Notice that "it is old
and well known to maximize the upsell of a product or service by providing
the customers with more expensive items, upgrades, add ons or additional
items in order to maximize the profits for the sellers." Id. The Examiner
cites to Walker, as support for the Official Notice, and finds that "Walker
teaches that up sells, cross sells, upgrades influence the purchase of the
products because they are related or products or services that work in
combination with other products purchased." Id.; see also Ans. 5---6 (citing
Walker i1 49). The Examiner determines that "[i]t would have been obvious
to a person taking into account purchasing influence of products on other
17
Appeal2017-001814
Application 13/312,511
products in order to obtain the above mentioned advantage." Id. at 8-9. The
Examiner further finds that "Paul teaches an influencer that can influence
others and Walker teaches an influence that a product has on another product
[and] Levy teaches in general combining all recommendations for blended
recommendations." Id. at 9 (citing Levy i-f 22). And, the Examiner
determines that one of ordinary skill in the art would combine
Id.
the general notion of combining different recommendations as
taught by Levy with relationship that a person has on a
product/Paul and that a product has on a second product/Walker
in order to better select an influencer that can influence both
another person and a second product in order to provide a better
blended recommendation.
We agree with the Appellants that "the claim limitation is not just
'maximiz[ing] the upsell of a product or service by providing the customers
with more expensive items, upgrades, add ons or additional items in order to
maximize the profits for the sellers,"' (Appeal Br. 24 (quoting Final Act. 8)
(alteration in original)), but requires an assessment of the first product's
correlation to or influence on the second, upsell product (id. at 26). The
Examiner does not adequately explain, and we do not see, that Walker
discloses assessing a first product's correlation to or influence on purchasing
the second offered product. The cited portion of Walker discloses upselling
an additional product that "supplements the customer's purchase." Walker
i-f 49. Walker provides examples of this upselling that include a restaurant
offering a promotional cup or dessert; a video store offering a movie rental
ticket, videotape, or hat; a vending machine offering another candy bar; an
appliance store offering a warranty; and a supermarket offering an item from
a bin of perishable goods. Id. At best, Walker discloses that the type of
18
Appeal2017-001814
Application 13/312,511
establishment correlates to or influences a second product. Further, as the
Appellants point out, Walker's second, upsell product is identified based on
the prices of the first and second products to reach a rounded price. Appeal
Br. 26 (citing Walkeri-f 51). But, the Examiner does not rely on or explain
how Walker teaches assessing the influence of the price of the first product,
as opposed to the rounded price, on purchasing the second product.
Similarly, the Examiner does not adequately explain how the selection of a
seed to maximize influence is determined based on the assessed influence of
the first product on purchasing the second product.
Based on the foregoing, we do not sustain the Examiner's rejection
under 35 U.S.C. § 103(a) of independent claims 13 and 14, and dependent
claims 15-17, 19, 22, and 25-29.
DECISION
The Examiner's rejection of claims 13-17, 19, 22, and 25-29 under
35 U.S.C. § 101 is AFFIRMED.
The Examiner's rejection of claims 13-17, 19, 22, and 25-29 under
the doctrine of non-statutory double patenting is REVERSED.
The Examiner's rejection of claims 13-17, 19, 22, and 25-29 under
35 U.S.C. § 103(a) is REVERSED.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l )(iv).
AFFIRMED
19
Application/Control No. Applicant(s)/Patent Under Patent
13/312,511 Appeal No. 2017-001814
Notice of References Cited Administrative Patent Judge 1 Art Unit
Amee A. Shah
Page 1 of 1
3600
U.S. PATENT DOCUMENTS
*
Document Number Date
Country Code-Number-Kind Code MM-YYYY Name Classification
A
B
c US-
D US-
E US-
F US-
G US-
H US-
I US-
J US-
K US-
L US-
M US-
FOREIGN PATENT DOCUMENTS
*
Document Number Date
Country Code-Number-Kind Code MM-YYYY Country Name Classification
N
0
p
Q
R
s
T
NON-PATENT DOCUMENTS
* Include as applicable: Author, Title Date, Publisher, Edition or Volume, Pertinent Pages)
Wibke Heidig, Upselling or Upsetting? Studies on the Behavioral Consequences of Upsell Offers in Service
u Encounters (Oct. 29, 2012) (unpublished Ph.D dissertation, University of St. Gallen) (available at
https://www1 .unisg.ch/www/edis.nsf/SysLkpByldentifier/4086/$FILE/dis4086.pdf)
v Larry Levine, Why Cross-Selling and Upselling Seem So Difficult to Implement, Telemarketing & Call Center
Solutions, Jan. 1, 1996 (HighBeam Research)
w
x
*A copy of this reference 1s being furnished with the associated Board dec1s1on from this appeal..
Dates in MM-YYYY format are publication dates. Classifications may be US or foreign.
U.S. Patent and Trademark Office
PT0-892 (Rev. 01-2001) Notice of References Cited Part of Paper No.
Upselling or Upsetting?
Studies on the Behavioral Consequences of Upsell Offers in
Service Encounters
DISSERTATION
of the University of St. Gallen,
School of Management,
Economics, Law, Social Sciences
and International Affairs
to obtain the title of
Doctor of Philosophy in Management
submitted by
Wibke Heidig
from
Germany
Approved on the application of
Prof. Dr. Torsten Tomczak
and
Prof. Dr. Andreas Herrmann
Dissertation no. 4086
Rosch-Buch, Schesslitz 2012
The University of St. Gallen, School of Management, Economics, Law, Social
Sciences and International Affairs hereby consents to the printing of the present
dissertation, without hereby expressing any opinion on the views herein expressed.
St. Gallen, October 29, 2012
The President:
Prof. Dr. Thomas Bieger
Fiir Marcus, Sigrid und Knut

Vorwort
Es waren Alltagsbeobachtungen, die mich zu meinem Dissertationsthema brachten.
Ich stellte mir die Frage, warum Kunden - mich eingeschlossen - auf das Angebot
einer hoherwertigen aber teureren Leistung manchmal positiv und manchmal
ablehnend reagieren. Diese Frage hatte es mir angetan, und so machte ich mich auf die
Suche nach Antworten. An dieser Stelle mochte ich nun all jenen Personen danken, die
mich auf diesem Weg begleitet und unterstiitzt haben.
Dank gebiihrt in erster Linie meinem Doktorvater Prof Dr. Torsten Tomczak fiir die
uneingeschrankte Unterstiitzung meines Dissertationsvorhabens und das motivierende
Vertrauen in meine Leistung. Seine fachliche und personliche Unterstiitzung sowie die
Moglichkeit, als wissenschaftliche Mitarbeiterin an der Forschungsstelle fiir Customer
Insight den akademischen und praktischen F okus zu vereinen, haben das Gelingen
dieser Arbeit erst ermoglicht. Bedanken mochte ich mich ebenso bei Prof Dr. Andreas
Herrmann fiir die freundliche Obemahme des Ko-Referats. Ein besonderer Dank
gebiihrt Prof Dr. Daniel Wentzel, der als mein akademischer Mentor in zahlreichen
intensiven Gesprachen und konstruktiven Diskussionen das Entstehen dieser Arbeit
begleitet und mich in allen Belangen des Studiendesigns und der Positionierung des
Themas stets unterstiitzt hat.
Auf meinem Weg haben mich viele Kollegen begleitet, die in den vier Jahren in
St. Gallen gute Freunde geworden sind. Fiir die schone und bereichemde Zeit in- und
ausserhalb des Instituts mochte ich mich daher unter anderem bedanken bei Suleiman
Aryobsei, Philipp Schaifenberger, Emanuel de Bellis, Lucas Beck, Julia Krimgen,
Fabian Heuschele, Dennis Vogt, Michaela Csik, Sven Molner und Simon Brosamle.
Ein spezieller Dank geht an Dr. Benjamin von Walter fiir die bereichemden und
humorvollen drei Jahre gemeinsamer Biirozeit. Dariiber hinaus danke ich Dr. Antonia
Erz fiir die sorgfaltige Korrektur der Arbeit und ihre freundschaftliche Unterstiitzung.
Der grosste Dank geht schliesslich an drei ganz besondere Menschen, ohne die dieser
Weg nicht moglich gewesen ware. Ich danke meinen Eltem Sigrid und Knut fiir die
bedingungslose Unterstiitzung meiner Vorhaben, die Zuversicht und den Glauben in
meine Fahigkeiten. Schliesslich danke ich meinem Partner und besten Freund Marcus
Hohn, der nicht nur in meinem Experimentalvideo die Hauptrolle spielt. Mit seiner
Liebe und seinem Verstandnis hat er mir in jeder Phase meiner Dissertation zur Seite
gestanden. Von Herzen widme ich ihnen diese Arbeit.
St. Gallen, im Oktober 2012 Wibke Heidig

VII
Abstract
"Would you like to supersize this room?" In many industries, especially in the service
sector, consumers are offered the opportunity to revise their initial reservation decision
in return for a superior but more expensive product or service option. For instance,
when checking in at an airport or hotel counter, consumers are frequently encouraged
to reconsider their initial reservation and book a higher-quality but more costly seat or
room. This marketing technique is referred to as upselling. Even though upselling
attempts are commercially appealing for companies and beneficial for consumers,
there is little research on how consumers respond to upsell offers. However, practice
and anecdotal evidence suggest that it is not self-evident that the consumer decides in
favor of the upsell. As such, one important question that remains to be addressed is
what factors drive a consumer's decision in favor or against an upsell offer. While
research on sequential decision making and preference reversals may provide initial
insights into consumers' behavioral intentions, it does not explicitly examine when and
why consumers decide to accept and pay for an upsell offer.
Hence, the purpose of this dissertation is to develop a conceptual model of the
upselling decision process and to empirically clarify when and why consumers may
accept such offers. Drawing on findings from the effort-accuracy framework, goal
framing and decision justification theory, the present research develops a theoretical
model that argues that the upselling process is best conceptualized as a three-step
decision process. The model proposes that extensive initial cognitive effort
investments induce a lock-in situation that reduces the willingness to choose the upsell
and to pay for it and that this effect is moderated by the goal frame of the upsell offer.
In total, four experiments confirm the models' propositions and provide substantial
evidence for the mediating role of anticipated inaction regret and decision
justifiability. Furthermore, it is tested whether decision delegation to a surrogate
shopper such as a travel agent alters these relationships and whether need for
justification acts as a boundary condition. The model and research results have
important implications for the services and sales literature and suggest that antecedent
decisions and associated cognitive effort investments exert a strong influence on
subsequent decisions regarding acceptance or refusal of the upsell offer. Thereby, this
dissertation contributes to research on selling techniques and identifies managerially
important levers on how to successfully manage upsell offers in the service encounter.
VIII
Zusammenfassung
In vielen Branchen, allen voran m der Dienstleistungsindustrie, wird Kunden die
Moglichkeit eingeraumt, eine urspriinglich getroffene Service- oder Produkt-
entscheidung zu Gunsten einer besseren und teureren Leistung zu revidieren. Dieses
Vorgehen wird als Upselling bezeichnet. Von Upselling-Angeboten wird beispiels-
weise gesprochen, wenn Kunden am Hotelempfang oder am Flughafenschalter die
Moglichkeit erhalten, ihre Reservierungsentscheidung zu iiberdenken und gegen einen
Aufpreis ein hochwertigeres Hotelzimmer oder einen Sitzplatz in einer besseren
Servicekategorie zu wahlen. Trotz der finanziellen Bedeutung von Upselling fiir
Untemehmen und der Vorteilhaftigkeit fiir den Kunden, gibt es bisher nur wenige
empirische Untersuchungen dazu wie Kunden auf diese Angebote reagieren. Erste
praktische und theoretische Erkenntnisse verdeutlichen jedoch, dass ein Wechsel des
Kunden zur besseren und teureren Leistung nicht selbstverstandlich ist. Die bisherige
F orschung im Bereich von sequentiellen Entscheidungsprozessen und Entscheidungs-
revisionen erklart dieses Verhalten nur unzureichend. Unbeantwortet bleibt die Frage,
wann und warum sich Kunden bereit erklaren ein Upselling-Angebot anzunehmen und
den Aufpreis zu zahlen.
Die vorliegende Dissertation widmet sich dieser Problemstellung. Ziel der Arbeit ist
es, ein konzeptionelles Modell von Entscheidungsprozessen in Upselling-Situationen
zu entwickeln und empirisch zu untersuchen, unter welchen Bedingungen Kunden auf
Upselling-Angebote eingehen. Aufbauend auf Erkenntnissen der kognitiven
Aufwandsforschung und der "Framing"-Literatur wird ein dreistufiges Entscheidungs-
modell des Kunden entwickelt. Das Modell postuliert, dass die Bereitschaft des
Kunden, das Upselling-Angebot anzunehmen, von dem Ausmass des kognitiven
Aufwands auf der ersten Entscheidungsstufe determiniert wird. Die Starke dieses
Einflusses auf die finale Entscheidung wird wiederum vom Framing des Upselling-
Angebots beeinflusst. Das entwickelte Modell nimmt zudem an, dass die Obertragung
der Entscheidungsverantwortung auf eine andere Person die Zusammenhange
modifiziert. Vier Experimente bestatigen die abgeleiteten Hypothesen und zeigen, dass
antizipiertes Bedauem und die Moglichkeit, die Entscheidung zu rechtfertigen, diese
Prozesse mediieren. Das entwickelte Modell leistet einen wesentlichen Beitrag fiir die
Dienstleistungsliteratur und verdeutlicht, dass vorausgegangene Entscheidungen und
das Ausmass an kognitiven Investitionen einen starken Einfluss auf nachfolgende
Entscheidungen iiber Annahme oder Ablehnung eines Upselling-Angebots haben. Die
vorliegende Dissertation identifiziert damit wesentliche Treiber fiir den erfolgreichen
Einsatz von Upselling-Angeboten am Serviceschalter.
IX
Table of Contents
Abstract ........................................................................................................................ VII
Zusammenfassung ...................................................................................................... VIII
Table of Contents .......................................................................................................... IX
List of Figures ............................................................................................................ XIII
List of Tables ............................................................................................................. XIV
1 Introduction .................................................................................................... 1
1.1 Problem Orientation ........................................................................................... 1
1.2 Research Questions and Structure of the Dissertation ....................................... 4
2 Conceptual Background ............................................................................... 7
2.1 Upselling and the Underlying Decision Process ................................................ 7
2.1.1 Defining and Delineating the Concept.. ....................................................... 7
2.1.2 Decisions in Upselling Situations .............................................................. 11
2.2 Initial Decision: The Central Role of Cognitive Effort .................................... 13
2.2.1 The Effort-Accuracy Framework ............................................................... 13
2.2.2 Switching Behavior and Cognitive Lock-in ............................................... 15
2.3 The Upsell Offer: The Moderating Role of the Message's Goal Frame .......... 17
2.3.1 Introduction to Message Framing .............................................................. 18
2.3.2 A Goal Frame View ofUpsell Offers ........................................................ 19
2.4 The Final Decision: A Question of Decision Justification ............................... 23
2.5 Initial Decision Revisited: The Impact of a Surrogate Shopper ....................... 25
2.5.1 Surrogates in Services ................................................................................ 25
2.5.2 The Role of Perceived Effort ..................................................................... 27
2.6 Linking the Steps: The Conceptual Model.. ..................................................... 30
3 Hypotheses Development ............................................................................ 32
3.1 The Interactive Effect of Cognitive Effort and Message Framing ................... 32
3.2 Need for Justification: A Boundary Condition ................................................. 36
3.3 Testing the Impact of a Surrogate Shopper ...................................................... 39
x
4 Experimental Analyses ................................................................................ 41
4.1 Overview of Studies ......................................................................................... 41
4.2 Experiment 1: The Moderating Role of Goal Framing .................................... 44
4.2.1 Design, Participants, and Procedure ........................................................... 44
4.2.2 Materials and Manipulations ...................................................................... 45
4.2.2.1 Manipulating Cognitive Effort ............................................................. 45
4.2.2.2 Manipulating the Message's Goal Frame ............................................ 48
4.2.3 Selection of Measures ................................................................................ 50
4.2.3.1 Dependent Variables ............................................................................ 50
4.2.3.2 Covariates ............................................................................................. 50
4.2.3.3 Manipulation and Confound Checks ................................................... 52
4.2.4 Results ........................................................................................................ 54
4.2.4.1 Checks on Experimental Design .......................................................... 54
4.2.4.2 Hypotheses Testing .............................................................................. 55
4.2.5 Discussion .................................................................................................. 59
4.3 Experiment 2: Cognitive Effort, Goal Framing and the Underlying Process .. 60
4.3.1 Design, Participants, and Procedure ........................................................... 60
4.3.2 Materials and Manipulations ...................................................................... 61
4.3.2.1 Manipulating Cognitive Effort ............................................................. 61
4.3.2.2 Manipulating the Message's Goal Frame ............................................ 63
4.3.3 Selection of Measures ................................................................................ 65
4.3.3.1 Dependent Variables and Process Measures ........................................ 65
4.3.3.2 Covariates ............................................................................................. 66
4.3.3.3 Manipulation and Confound Checks ................................................... 67
4.3.4 Results ........................................................................................................ 68
4.3 .4.1 Checks on Experimental Design .......................................................... 68
4.3.4.2 Hypotheses Testing: Moderation ......................................................... 69
4.3.4.3 Hypotheses Testing: Mediation ........................................................... 72
4.3.5 Discussion .................................................................................................. 74
4.4 Experiment 3: Decision Justification and Goal Framing ................................. 76
4.4.1 Design, Participants, and Procedure ........................................................... 76
4.4.2 Materials and Manipulations ...................................................................... 77
4.4.2.1 Manipulating Need for Justification .................................................... 78
4.4.2.2 Manipulating the Message's Goal Frame ............................................ 79
4.4.3 Selection of Measures ................................................................................ 82
4.4.3.1 Dependent Variables and Process Measures ........................................ 82
XI
4.4.3.2 Covariates ............................................................................................. 83
4.4.3.3 Manipulation and Confound Checks ................................................... 83
4.4.4 Results ........................................................................................................ 85
4.4.4.1 Checks on Experimental Design .......................................................... 85
4.4.4.2 Hypotheses Testing: Moderation ......................................................... 86
4.4.4.3 Hypotheses Testing: Mediation ........................................................... 90
4.4.5 Discussion .................................................................................................. 92
4.5 Experiment 4: Perceived Effort and Goal Framing .......................................... 94
4.5.1 Design, Participants, and Procedure ........................................................... 94
4.5.2 Materials and Manipulations ...................................................................... 95
4.5.2.1 Manipulating Perceived Effort ............................................................. 96
4.5.2.2 Manipulating the Message's Goal Frame ............................................ 98
4.5.3 Selection of Measures ................................................................................ 98
4.5.3.1 Dependent Variable .............................................................................. 98
4.5.3.2 Covariates ............................................................................................. 98
4.5.3.3 Manipulation and Confound Checks ................................................... 99
4.5.4 Results ...................................................................................................... 100
4.5.4.1 Checks on Experimental Design ........................................................ 100
4.5.4.2 Hypothesis Testing ............................................................................. 101
4.5.5 Discussion ................................................................................................ 103
5 General Discussion .................................................................................... 105
5.1 Summary of Results ....................................................................................... 105
5.2 Theoretical Contributions ............................................................................... 109
5.2.1 Contribution to Literature on Upselling ................................................... 109
5 .2.2 Contribution to Research on Cognitive Effort ......................................... 110
5.2.3 Contribution to Research on Goal Framing ............................................. 111
5.2.4 Contribution to Research on Decision Justification ................................. 112
5.3 Managerial Contributions ............................................................................... 113
5.3.1 Implications for Communication Strategies ............................................. 113
5.3.2 Implications for Pricing ............................................................................ 115
5.3.3 Understanding Sequentiality and Justifiability ofUpsell Decisions ........ 115
5.4 Limitations ...................................................................................................... 116
5.5 Further Research and Extensions ................................................................... 118
5.5.1 Lock-in Factors ........................................................................................ 118
5.5.2 Contextual Factors of the Service Encounter. .......................................... 120
5.5.3 Macro-level Factors ofUpsell Offers ...................................................... 123
XII
6 References ................................................................................................... 125
7 Appendices .................................................................................................. 145
7 .1 Appendix 1: Stimulus Materials used in Experiment 1. ................................. 145
7.2 Appendix 2: Stimulus Materials used in Experiment 2 .................................. 148
7.3 Appendix 3: Stimulus Materials used in Experiment 3 (Car Rental) ............. 152
7.4 Appendix 4: Stimulus Materials used in Experiment 3 (Hotel) ..................... 155
7.5 Appendix 5: Stimulus Materials used in Experiment 4 .................................. 159
XIII
List of Figures
Figure 1-1: Structure of the Dissertation ..................................................................... 6
Figure 2-1: Conceptualization of Decision Processes in Upselling Situations .......... 12
Figure 2-2: Conceptual Model of the Dissertation .................................................... 31
Figure 4-1: Overview of Experimental Studies ......................................................... 43
Figure 4-2: Low Cognitive Effort Manipulation in Experiment 1 ............................ 47
Figure 4-3: High Cognitive Effort Manipulation in Experiment 1 ............................ 4 7
Figure 4-4: Cognitive Effort and Goal Framing Interactions (Experiment 1) ........... 58
Figure 4-5: Low Cognitive Effort Manipulation in Experiment 2 ............................ 62
Figure 4-6: High Cognitive Effort Manipulation in Experiment 2 ............................ 62
Figure 4-7: Snapshot of the Video Vignette .............................................................. 64
Figure 4-8: Cognitive Effort and Goal Framing Interactions (Experiment 2) ........... 71
Figure 4-9: Mediation Models in Experiment 2 ........................................................ 73
Figure 4-10: Manipulation of Independent Variables -Car Rental (Part I) ................ 80
Figure 4-11: Manipulation of Independent Variables - Car Rental (Part II) .............. 81
Figure 4-12: Choice Share of the Upsell Option (Experiment 3) ................................ 86
Figure 4-13: Need for Justification and Goal Framing Interactions (Experiment 3) .. 89
Figure 4-14: Mediation of Anticipated Inaction Regret in Experiment 3 .................... 91
Figure 4-15: Mediation of Decision Justifiability in Experiment 3 ............................. 92
Figure 4-16: Perceived Effort and Goal Framing Interaction (Experiment 4) .......... 102
XIV
List of Tables
Table 2-1: Empirical Studies on Upselling .............................................................. 10
Table 4-1: Measures employed in Experiment 1 ...................................................... 54
Table 4-2: Results of the ANCOV As in Experiment 1 ............................................ 57
Table 4-3: Mean Values for the Dependent Variables in Experiment 1 .................. 57
Table 4-4: Measures employed in Experiment 2 ...................................................... 68
Table 4-5: Results of the ANCOV As in Experiment 2 ............................................ 70
Table 4-6: Mean Values for the Dependent Variables in Experiment 2 .................. 70
Table 4-7: Measures employed in Experiment 3 ...................................................... 84
Table 4-8: Results of the ANOV As in Experiment 3 ............................................... 88
Table 4-9: Mean Values for the Dependent Variables in Experiment 3 .................. 88
Table 4-10: Measures employed in Experiment 4 .................................................... 100
Table 4-11: Results of the ANCOVA in Experiment 4 ............................................ 102
Table 4-12: Mean Values for the Dependent Variables in Experiment 4 ................ 102
1 Introduction
1.1 Problem Orientation
Imagine the following situation: Spring is in the air - and summer is on
the way- and you decide to spend this year's summer vacation visiting
one of the most vibrant cities in Europe - Berlin. Since you love to
travel by car instead of flying long distances, you drive the journey with
your SUV. Now, the only thing you have to arrange in advance is to
reserve a hotel room. Clicking through numerous hotel and broker
websites you finally find a hotel that is centrally located next to most of
Berlin's historic sites that you do not want to miss on your city trip. The
website of your preferred hotel offers you a wide range of vacant
rooms, differing in size, equipment, and service. For every available
room you find an extensive description of all amenities it has to offer.
Accordingly, it takes you some time to evaluate all the options, to weigh
the pros and cons, and to find the room you finally want to reserve for
your vacation. But you finally make it. You choose a single bed room
located at the backside of the hotel, press the "Reserve" button, and
look forward to this thrilling city trip.
A few weeks later the time has come - your holiday starts. After a few
hours' drive you arrive in Berlin. You directly go to the hotel you
placed your reservation at in order to check-in and pick up your room
key. Approaching the hotel counter, you hand your reservation
confirmation to the desk clerk. The attentive receptionist checks your
reservation details, points at your bags and says: "The reserved room
is ready for you, but let me call your attention to a special offer. For a
small daily extra fee I can offer you a room with more amenities, a
king-sized bed, and a delightful view instead. The room that I can offer
you is spacious and comfortable enough to store your luggage. "
Now you start to reflect on this offer. How should you decide? Should
you opt for the superior room or rather stick to the room you had
initially reserved?
2
This scenano is a realistic example of a situation that occurs in many service
encounters every day. A customer who has already decided on a product or service
option is confronted with an enhanced version of that product or service in turn for a
small premium. In this situation, the customer has to decide whether to accept the
enhanced product or service offer and thus to reconsider the initial choice or to
continue with the initial reservation. The employed selling technique is referred to as
upselling.
Upselling is an established and widely utilized sales tool especially in the travel
industry (cf. Kamakura, 2008). It denotes a salesperson's attempt to persuade "a
customer to purchase a higher-level product or service, richer in functions for the user
and more profitable for the company" (Vercellis, 2009, p. 335). Due to its capability to
exploit finite capacities more profitably, upselling is a common practice for service
providers that offer different service classes such as airlines (cf. Bohutinsky, 1990),
hotels (cf. Lewis, 2005), car rentals (cf. Goldstein, 1997) and performing arts
companies (cf. Biyalogorsky et al., 2005). Especially in these industries, demand
fluctuations and perishable inventories (Wirtz et al., 2003) heighten the probability
that some upscale units like rooms on the executive floor or prestige cars might be left
unsold although they are available and would fit the customers' needs (Biyalogorsky et
al., 2005). In these situations, offering upper-class units to customers that have already
reserved low-grade services in advance seems beneficial for both - the service
provider as well as the customer.
Hence, upselling bears a non-neglectable financial potential for service firms because
it allows them to flexibly increase inventory utilization during low demand periods. It
is especially important for a firm's revenue because selling superior products or
services at higher prices increases profits at equal costs. The costs remain almost
equal, because of a fixed inventory and a low variable to fixed costs ratio (Wirtz et al.,
2003). This can be easily exemplified with an abbreviated calculation of a hotel room
(see e.g., Bohutinsky, 1990): Assuming that you reserved a single bed room for a daily
room rate of Ra. With total costs C of assisting and cleaning the room this generates a
profit Pa for the hotel, calculated as the difference between the room rate and the costs.
Now imagine you were willing to opt for the superior king-sized room with a higher
room rate Rb and a new profit Pb. Assuming that the costs remain constant due to fixed
resources, this generates an increase in profit (Pb- Pa)· The economic potential of upsell
offers is therefore obvious.
In addition, upsell offers also benefit customers since they compensate for customers'
limited ability to construe distant events. Research on construal level theory (CLT)
3
found that customers find it difficult to anticipate future preferences (Trope &
Liberman, 2000, 2010). As such, customers may fail to predict which product or
service option ideally suits their preferences when the moment of reservation and the
moment of consumption are temporally departed. Accordingly, a reservation for a
vacation in six months will be driven by reflections on essential features rather than a
reservation for a hotel stay this weekend where more incidental features become
important (Fiedler, 2007; Trope & Liberman, 2000). For example, a customer1,
booking a hotel room for the summer holidays does not inevitably consider the number
of bags and the associated storage space. Consequently, exposing customers to
products or services that have not been considered for reservation but satisfy current
needs more effectively are undoubtedly beneficial for customers.
Despite the prevalence and commercial appeal of upselling attempts in the service
industry, research on the determinants of successful upsell options remains limited.
This may be at least partially due to the vague conceptualization and fuzzy
discrimination of upselling from related selling activities such as cross-selling and
upgrading, as well as the offering of add-ons (Pohlkamp, 2009). In fact, only few
publications build upon a distinct definition and conceptualization of upselling. These
publications investigate two different aspects. First, a number of studies can be
identified that explore upselling in the light of the organizational behavior and
motivation literature. They investigate the drivers of an increased upsell performance
of employees (Squires et al., 2007; Wiesman, 2006). Second, under the rubrics of up-
and downward migration, several publications discuss the upselling concept within the
context of customer relationship management and customer lifetime value (Bolton et
al., 2008; Kim & Kim, 1999; Ngobo, 2005; Pohlkamp, 2009; Salazar et al., 2007).
Based on segmentation approaches, these studies analyze the upselling potential within
ongoing relationships between a firm and its customers. This research stream views
upselling as a strategic means for relationship expansion and largely explains a
customer up-buying decision by means of social demographics (Bolton et al., 2008).
However, research on the psychological drivers of the customers' decisions in the
service encounter2 remains scarce. No research attempts have been made in order to
investigate the critical factors and effects ofupsell offers in the direct person-to-person
interaction between the customer and the employee of the service provider without
1 The terms "customer" and "consumer" are used interchangeably in this dissertation in order to include served as well as
potential customers.
2 According to Surprenant and Solomon (1987), a service encounter can be defined as "the dyadic interaction between a
customer and a service provider" (p. 87). While adapting this definition, the current research acknowledges the interactive
nature of service encounters that arises from personal interactions between the customer and the service employee. For
further discussions on the conceptual boundaries see Bitner et al. (1990).
4
assuming a prior business relationship. This lack of research is surprising considering
that recent publications anecdotally highlight that successful upselling requires an
elaborated understanding of a firms' customers (Salazar et al., 2007; Wirtz et al.,
2003). Furthermore, practice shows that it is not self-evident that the customer decides
in favor of the upsell offer although it might be obviously beneficial. In contrast, it
might also be detrimental for future sales of the company if the customer feels
pressurized or overwhelmed (see e.g., Teague, 2011). Once a customer has decided on
a product or service, he or she is often unwilling to opt for another product or service
option. Hence, it seems worthwhile to investigate those factors that drive a customer's
willingness to choose and additionally pay for the upsell option. This dissertation will
account for the multi-level nature of those decision processes that involve a preceding
reservation decision and a subsequent upsell offer. Thus, the present dissertation will
disclose variables that determine the upselling success at each of the decision steps.
1.2 Research Questions and Structure of the Dissertation
The purpose of this dissertation is to investigate how and why consumers respond to
upsell offers and to identify variables that drive their decision processes. As such, the
overall objective of this dissertation is to conceptually analyze the decision process
that underlies an upsell choice and to empirically examine the influence of, and
interplay between, different decision variables at each of the decision steps. Research
on selling strategies has largely failed to account for the influence of prior experiences
and decisions on subsequent decisions at the service counter. Consequently the current
research addresses this void by proposing a three-step choice process that is assumed
to apply for those decision contexts where upselling is routinely employed - namely in
the service industry. It is assumed that customers commit themselves to a specific
product or service option at an initial reservation step. This commitment is likely to be
put to test at a later decision step when the salesperson offers an upsell to the
customer. In order to determine the influence of the initial decisions on customers'
reactions to a subsequent upsell offer, a conceptual model is proposed that draws upon
effort-accuracy research, framing literature and insight from decision justification
theory. To ascertain a thorough conceptual and empirical investigation, this model will
incorporate a number of moderating variables that are assumed to be theoretically
important and managerially relevant. Moreover, it will also specify mediation
processes and conditions under which the influence of an initial decision will be
reinforced or attenuated. In total, four studies will test the proposed model in order to
answer the following research questions:
5
Research Question 1:
How and why does the initial product or service choice of the customer
influence the subsequent decision of whether to accept or to refuse the
upsell offer? More specifically, to what extent does the initial cognitive
effort investment influence the upsell choice?
Research Question 2:
Are different selling arguments likely to moderate the impact of the initial
product or service choice on the subsequent decision? Put differently, which
message frame attenuates or intensifies the influence of the initial cognitive
effort investment on the upsell choice?
Research Question 3:
Which process variables, i.e., mediators, explain the customer's choice or
refusal of the upsell offer?
Research Question 4:
How does the decision process change when the customer delegates the
initial reservation decision to a surrogate shopper? More precisely, is the
cognitive effort exerted by a surrogate likely to act as a moderator?
These research questions will guide the following theoretical and empirical
investigations that are structured into five main chapters. Chapter 1 gives an
introduction to the dissertation's topic and depicts the research questions. Chapter 2
develops a theoretical foundation for this dissertation. It introduces the upselling
concept as well as the underlying decision process in greater depth. For the most part,
this chapter reviews the relevant literature on those variables, which are assumed to
influence the customer's decision at each decision step. Based on the literature on
cognitive effort, message framing and decision justification this chapter finally
concludes with a conceptual model. Following in Chapter 3, a number of research
hypotheses are developed and put to test in Chapter 4. That is, the fourth chapter
presents the findings from four experimental studies that differ in terms of research
design and focus. For each experiment, an extensive documentation will be presented,
including details on design, method and results.3 Finally, Chapter 5 summarizes the
findings of all studies and addresses the key theoretical and managerial contributions
and limitations. The dissertation concludes with a prospect on promising avenues for
future studies. Figure 1-1 provides an overview of this structure.
3 The structure of the fourth chapter follows the outline employed by Wentzel (2008).
6
7
2 Conceptual Background
The purpose of this chapter is to present a comprehensive review of the theoretical
foundation of this dissertation. It provides a conceptual clarification of the customer's
decision process that underlies an upsell choice and reviews the relevant literature on
those variables that may affect a customer's final decision regarding acceptance or
rejection of the upsell offer. The first part of this chapter will introduce the concept of
upselling and a model of the associated decision steps. This model provides a frame of
reference as the following sections are organized along these decision steps. They will
explore those variables and theories that are expected to affect a customer's
willingness to accept the upsell offer and to pay an extra charge. More precisely, the
next four paragraphs review the extant literature on cognitive effort, goal framing,
decision justification and surrogacy. The last part of this chapter integrates the
theoretical reflections in a unifying conceptual model and serves as a basis for the
hypotheses formulation in Chapter 3.
2.1 Upselling and the Underlying Decision Process
2.1.1 Defining and Delineating the Concept
Upselling is an established and widely utilized sales tool employed by service
companies in order to increase the purchasing value of their customers (Kamakura,
2008). Thereby, the company attempts to persuade the customer "to purchase a higher-
level product or service, richer in functions for the user and more profitable for the
company" (Vercellis, 2009, p. 335). Despite the commercial appeal and practical
prevalence of upselling it has not been widely studied yet. This might be due to two
reasons. First, researchers and practitioners do not agree on how to conceptualize
upselling. While some authors indicate that upselling is best defined as a strategic
instrument of customer relationship management (e.g., Bauer et al., 2003; Tomczak et
al., 2009) other authors consider upselling as an operative selling technique employed
in the direct service encounter (e.g., Biyalogorsky et al., 2005). Second, due to its
fuzzy delineation from related selling activities such as cross-selling and upgrading as
well as the offering of add-ons, publications lack in conceptual clarity (e.g., Squires et
al., 2007). Consequently, the following paragraphs will distinguish the different
concepts and provide a unifying definition and foundation for this dissertation.
8
The few publications dealing with upselling almost exclusively focus on its strategic
application as an instrument to increase the lifetime value of customers in ongoing
long-term client-business relationships (Bauer et al., 2003). Based on segmentation
approaches these studies calculate the upselling potential for different industries such
as life insurances (Kim & Kim, 1999) and railway companies (Pohlkamp, 2009). From
this perspective, upselling is a valuable tool to enhance the revenue from long-term
customers that have already bought products or services from the provider before. For
example, car manufacturers like BMW offer a product portfolio with various vehicle
sizes in order to upsell customers of entry-level models like the BMW 1 to superior
models in the course of the customer lifetime cycle. Since long-term customers are less
price-sensitive and more knowledgeable of the product line offered by the company,
they are prone to increase their purchase amount or purchase value over time (Reinartz
& Kumar, 2000). Therefore, strategic upselling is a common practice in the consumer
durables and service industry (Pohlkamp, 2009).
Besides, upselling as used in this dissertation, can also be conceptualized as an
operative selling technique without assuming prior customer-seller relationships. It
provides the capability to manage capacities profitably and is therefore an integral part
of a firm's revenue (or yield) management (Talluri & van Ryzin, 2004). Behind other
instruments such as price fencing and overbooking, upselling is a suitable tool to
allocate fixed resources flexibly to different customer segments (Wirtz et al., 2003).
Particularly in the tourism (airlines, car rentals, hotels, cruises) and performing arts
industry demand uncertainties and no-shows4 heighten the probability that some
upscale service class units might be left unsold (Biyalogorsky et al., 2005).
Inventories, such as aircraft seats or rental cars are perishable in that, for example,
revenue is lost during non-usage (Subramanian et al., 1999; Wirtz et al., 2003).
Therefore, upselling customers to upper-class services balances demand fluctuations.
Moreover, professionally employed upselling also benefits customers for the following
reasons. First, offering products or services with enhanced functionality or comfort
right before usage or consumption allows customers to adjust their decision according
to their latest preferences and needs. In general, customers find it difficult to anticipate
which product or service features might become important in the specific moment of
consumption (Trope & Liberman, 2000, 2010). While reserving a service in advance
they are likely to choose those options that are attractive in the short term. They tend
to neglect those options that would yield future benefits and underestimate the impact
4 A "no-show" describes a person who reserves a space, e.g., on an airplane, in a theatre or in a hotel but neither cancels nor
uses the reservation.
9
of future contextual factors (Zauberman, 2003). Thus, offering additional and
advanced products or services in the service encounter balances the missing ability to
construe future events. Second, some product or service classes that may not have
been available at the time of reservation might become available right before
consumption or usage. Accordingly, adding these options to the choice set benefits
customers while increasing variety.5 These arguments show that, on the one hand,
customers may benefit from upsell opportunities and on the other hand customers play
a critical role for the success of an upsell offer. Although this notion is not new to the
literature on revenue management (Salazar et al., 2007), research in this area almost
exclusively focuses on forecasting and optimization models (Wirtz et al., 2003). To
date, only two studies can be identified that examine the drivers of upselling decisions
from a consumer behavior perspective. A study conducted by Liu (2008; study 4)
examines upselling in the context of decision interruptions. She found that participants
who had to decide whether to accept or to refuse an upsell offer were less price
sensitive when their decision process was interrupted. In a similar vein, Yi and
Baumgartner (2008) examined the influence of motivational compatibility between the
chronic regulatory focus of a customer and the frame of the upsell message. However,
both studies used upselling as sample applications of their proposed hypotheses.
Upselling was not the focal construct of their research. Accordingly, their research
does not account for the multi-level nature of upselling processes (see Section 2.1.2)
and leaves much space for further research.
Table 2-1 summarizes the key findings of the existing studies dealing with upselling
either from a strategic perspective of customer relationship management or from an
operative view of upselling as a tool of a firm's revenue management. The present
dissertation clearly contributes to the latter. In addition, there are a number of
publications that were not incorporated in this list because of a lack of differentiation
towards related concepts such as cross-selling and upgrading. While these concepts
share some common elements with upselling, they are quite distinct from each other in
terms of the underlying decision process and the customer's required willingness to
pay. For these reasons it appears useful to conceptually distinguish between these
constructs and to provide a clear definition ofupselling as it is used in this dissertation.
5 Because of this characteristic, upselling as it is commonly employed in practice and used in this dissertation is distinct from
unlawful ''bait and switch" tactics that attempt to deceive customers (Biyalogorsky et al., 2005). Bait-and-switch techniques
typically involve a "bait" aid that calls the customers' attention and prompts them to visit the store. The customer is finally
forced to "switch" to another, higher priced product because the original offer is no longer available (Wilkie et al., 1998).
Upselling in the sense of this dissertation simply implies that the salesperson attempts to persuade the customer to choose a
higher level product. Because the originally reserved product or service option is still available, the customer can finally
choose between both offers.
10
Table 2-1: Empirical Studies on Upselling •••. ,, ••••••••••••• ~"'\:~~ ••••.•••••••••••••••••••••• ~,~ ••••••••••••••••• Md. -
•II I Kim. - (1'99)
I II Ngobo (2005)
l~Jr ltl 11 1-~1 Salazar et al. (2007)
• Industry: life insurance Context specific findings: Identification of an upselling
• Identification of upselling potential potential of an additional 25% for more than half of the
and introduction of a methodology customers. Type of profession and age act as predictors of
to calculate upselling potential upselling potential.(= segmentation approach)
• Industry: performing arts
• Identification of reasons for upward
(subscribers) and downward
(occasional ticket buyers) migration
of customers
• Industry: financial services
• Identification of cross·selling and
upselling opportunities to reinforce
customer retention in ongoing
relationships
The combination of positive experiences with individual
(age, gender, education) and relational (frequency of
interaction) characteristics drive upward migration. No
relationship can be found between service satisfaction and
downward migration. (""segmentation approach)
The lifecycle stages of the customers determine their
consumption of cross· and upselling offers. These stages
encompass variables such as age and previous experiences
with the company.("" segmentation approach) 1~1 t\I l~.1 • Industry: industrial services A model-estimation with more than 2000 service contracts
1'..-M@mJ@M Bolton et al. (
2
00S) • Identification of variables that reveals that satisfaction, service quality, and price directly
~-------·mD determine a business customer's influence the upsell decision of business decision makers,
W:c~ decision to expand service contracts while price and satisfaction also moderate the influence of ®iii instead of renewing an existing one service quality. ( = model estimation)
•••• • Industry: railway company Age, aspiration level, duration of the relationship as well as
•••••••
Pohlkamp (
2009
) • Identification of the upsetling the amount of collected bonus points discriminate between
potentials of different customer customers who choose the upsell and customers who do
."""""""""""""""""""'""""":~:::~~""""""""""""""""""""""""":::,~~'~'~:~::~::,::~,~,~~~""""""""""""""""""""""""" •• 1~1 Liu (2008)
II II 1011 1_1@_1 Yi & Baumgartner
1~1 (2008)
1;~1 ~
• Industry: car rental The research does not focus on upselling choices itself but
• Examines the influence of decision on decisions associated with decision interruptions. It
interruptions on price sensitivity in shows that interrupting upselling decisions decreases price
upselling decisions sensitivity (study 4). ("' e:gperiment)
• Industry: airline Persuasion of the upsell message is enhanced when the
• Examines the interactive effect of focus of the message is compatible with the decision
framing and regulatory focus of the maker's motivational system, i.e., his or her chronic
message on persuasion regulatory focus (study 1 ). (""experiment)
First, cross-selling can be defined as the offering of complementary or unrelated
additional products or services (so called add-ons) to a previously chosen or purchased
product or service (Bertini et al., 2009; Gabaix & Laibson, 2006; Kamakura, 2008; Li
et al., 2011). These products or services expend the initially chosen option and require
the customer to hold a certain willingness to pay. When the customer decides in favor
of the cross-sell option, this product or service is in addition to the initially chosen
option - the customer does not need to reconsider a choice once made. A trip
cancellation insurance that airlines often offer their customers at the end of their
booking process is a prominent example of a cross-sell offer. Second, with an upgrade
a customer receives more service than initially expected without side-paying for the
increased output. This "overfulfilment of the service promise" results from
overbooked capacities in one or more service classes (von Wangenheim & Bayon,
2007, p. 3 7). Upgrading is similar to upselling in that the customer is offered a product
11
or service that substitutes the initially chosen product or service option. In contrast to
an upsell, an upgrade offer does not require the customer to pay for the enhancement
(Pohlkamp, 2009). Upgrading guests from the economy class to a seat in the business
class is a common procedure especially in the airline industry. Third, an upsell offer
requires the customer to reconsider the initially reserved product and service option. It
is distinct from cross-selling in that upselling substitutes an already chosen product or
service and it differs from upgrading in that the customer needs to pay an extra charge
(Pohlkamp, 2009). Consequently, upselling in the sense of this dissertation can be
defined as follows:
Upselling describes a salesperson's attempt to persuade a customer to choose a
superior and more expensive product or service option, richer in functions for the
customer and more profitable for the company instead of the initially chosen product
or service option.
2.1.2 Decisions in Upselling Situations
In light of this definition, it is crucial to illustrate the typical choice process behind an
upsell choice. Compared to a cross-sell offer that encourages a customer to add an
additional product or service to the initial choice option, an upsell offer prompts the
customer to reconsider the initial decision und therefore involves a quite unique
decision pattern. Hence, one can assume that the decision process behind an upsell
choice typically encompasses three steps. 6 At a first step, the decision maker chooses
from a given set of available options according to his or her preferences. This decision
can be either done by the customer or by a decision surrogate like a travel agent or
secretary. While the customer chooses for his or her own purposes, the surrogate
decides on behalf of the customer. Recalling the hotel example, this implies that the
customer or surrogate chooses a room from a set of available options within a certain
price category. In the context of services, this decision most often results in a
reservation. Since the first step does not inevitably take place in the direct service
encounter, there may be a greater or lesser time delay towards the second step of the
process. The second step denotes the direct customer-seller interaction in the service
encounter in which the salesperson submits the upsell offer to the decision maker. The
final decision in favor of or against the upsell is the third step of the process. Figure
2-1 depicts the conception of a prototypical decision process in upselling situations.
6 The three-step decision process cannot only be derived from the different conceptualizations and definitions of upselling but
also from the upselling practice. At a preliminary research stage of this dissertation, several interviews with representatives
from the hotel and car rental industry have been conducted in order to validate the concept.
12
Figure 2-1: Conceptualization of Decision Processes in Upselling Situations
~'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-');
~ ~
~ ~
~ Personal (Reservation) !
~ . . ~
~ Dec1S1on ~
~ ~
~ ~
~'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''~
~'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''')\
~ ~
~ ~ ! Surrogate (Reservation) !
~ Decision ~
~ ~
~ ~
~ ..... ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ... ~
"'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''\:
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~ ! Upsell Offer of the !
~ Salesperson ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
t,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ... ,~
"''-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-"'\:
l ' ! Acceptance of the Upsell !
~ Offer ~
' l l '
~ (Rejection of the Initial !
~ Choice) ~
' l ~ ~
~ .... ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,~
~''''''''''''''''''''''''''''''''''''''''''''''''''''''''1
~ Rejection of the Upsell ~
~ ~
~ Offer ~
~ ~
~ (A f ~ ~ cceptance o the ~
~ Initial Choice) ~ ~ ~
~''''''''''''''''''''''''''''''''''''''''''''''''''''''''~
The figure shows that decision processes in upselling situations emerge as
intertemporal combinations of two choices - the reservation decision and the final
decision on the upsell offer. There is evidence to assume that the initial reservation
decision impacts the final decision of the customer. In the last years, research on
sequential decision making has begun to investigate how preceding decisions and
behaviors influence subsequent choices and intentions (e.g., Albarracin & Wyer, 2000;
Bechwati & Siegal, 2005; Khan & Dhar, 2006; Kidwell & Jewell, 2008; Novemsky &
Dhar, 2005; Ofek et al., 2007). In a model incorporating bounded rationality, Ofek et
al. (2007) investigated how recall of information from prior decisions affected
willingness to pay for a subsequent decision. It showed that decision makers use
information from prior decisions to temper uncertainty about option utility in current
decisions. In a similar vein, Novemsky and Dhar (2005) argue that prior decisions
serve as a frame of reference for subsequent decisions. These results allow for the
general conclusion that the initial reservation decision and the associated information
impact the final choice. If this was the case, however, the question arises as to which
specific information from the initial reservation decision is most likely to impact the
decision on whether to accept or to reject the upsell offer. In a study on technological
product enhancements Okada (2001, 2006) found that prior product choices influenced
subsequent replacement decisions via the monetary price spent on the initial product. It
seems that customers mentally track financial investments over time and that these
investments influence prospective purchase decisions (Heath & Soll, 1996). A
practically relevant and realistic assumption of the present research is that the initial
13
reservation decision does not involve any financial investments. The customer is
supposed to pay at a later decision stage. Hence, financial effort may not account for
the final decision. Instead, it seems most likely that the cognitive effort investment in
the initial reservation decision impacts the final decision. So far, research on sequential
decision making has been largely silent on the influence of prior cognitive effort
investments on subsequent decisions. Therefore, the next section will review studies in
the field of cognitive effort that may provide valuable insights for the current research.
2.2 Initial Decision: The Central Role of Cognitive Effort
In order to assess how initial cognitive effort investments impact upsell decisions, a
general conception of cognitive effort and its role in decision making has to be
developed. In general, effort includes "the physical, mental, and financial resources
expended to obtain a product" (Cardozo, 1965, p. 244). The construct of cognitive
effort is of major importance in behavioral decision research (Bechwati & Xia, 2003;
Bettman et al., 1990; Russo & Dosher, 1983) and cognitive psychology (Kahneman,
1973). Cognitive effort can be defined as the "total use of cognitive resources required
to complete the task" (Payne et al., 1990, p. 298). It encompasses perception and
information acquisition, judgment and computational strain (Bettman et al., 1991;
Cooper-Martin, 1994) and is subjectively experienced as "trying hard" (Yeo & Neal,
2008, p. 617). In what follows, the next chapters will focus on two concepts, the effort-
accuracy framework and switching costs, which may provide an answer to the
question of how cognitive effort affects decision processes.
2.2.1 The Effort-Accuracy Framework
Empirical (Bettman et al., 1990; Fennema & Kleinmuntz, 1995; Russo & Dosher,
1983), simulative (Johnson & Payne, 1985; Payne et al., 1990) as well as conceptual
(Kleinmuntz & Schkade, 1993) research reveals that cognitive effort largely
determines the strategy selection of decision makers and acts as a determinant of
human decision making (Bettman et al., 1990). This robust finding is reflected in the
effort-accuracy framework (Bettman et al., 1993; Johnson & Payne, 1985; Payne,
1982), a descriptive scheme of a decision maker's behavior (Bechwati & Xia, 2003).
According to this framework, a decision maker strives to maintain a balance between
the effort invested in a decision strategy and the accuracy of choice. Put differently,
the decision maker trades off benefits and costs against each other in order to select a
decision rule (Bechwati & Xia, 2003). While cognitive effort is the major component
14
of cost and accuracy of choice is the principal component of benefit, the framework is
often referred to as the cost-benefit framework (Bettman et al., 1991; Shugan, 1980).
The general assumption underlying this framework is that the decision maker tries to
minimize effort and maximize accuracy (Bechwati & Xia, 2003; Bettman et al., 1991;
Payne, 1982). As a "cognitive miser" (Fiske & Taylor, 1984, p. 12), the decision
maker chooses the choice strategy according to the anticipated benefits of the strategy
as well as the cognitive effort involved with this strategy (Payne & Bettman, 1992).
According to Payne ( 1982), costs and benefits are not absolute but depend upon the
characteristics of the task, the context and the decision maker as well. Previous
experiences and chronic foci (e.g., attitude towards risk), on the one hand, provide
examples of such individual characteristics of the decision maker (Iglesias-Parro et al.,
2001). They typically affect accuracy and effort equally. Context variables, on the
other hand, primarily affect the accuracy of the decision because they influence the
values of the choice alternatives through e.g., relative similarity or dominance
structures (Bettman et al., 1993). In contrast, task variables exert their strongest effect
on the relative effort needed to execute the decision strategy (Bettman et al., 1993).
They are of key importance for the current research.
Task variables influence the amount of cognitive effort by either affecting the
resources required to manage the task or by affecting the individual resources available
(Garbarino & Edell, 1997). Factors such as time constraints or distraction influence the
available resources of the decision maker (Bettman et al., 1993; Garbarino & Edell,
1997). Task factors that extend the required resources are, for example, the number of
attributes, the novelty of the decision task, the ambiguity of the decision goal, the
complexity of the decision problem, the irreversibility or the significance of the
decision outcomes (Garbarino & Edell, 1997; Payne, 1982). It is exactly those task
variables like complexity and novelty of the decision task that are most likely to
influence the amount of cognitive effort a customer has to invest in a reservation
decision. This is because the reservation systems of service providers differ in terms of
complexity and presentation format. Given the multi-optional character of service
categories like, for example, room categories in a hotel, the customer often has to exert
a high amount of effort in order to assess and to compare the different service
alternatives and to come to an initial reservation decision.
Consequently, what happens to the decision maker and his final decision on the upsell
offer if the initial reservation decision required a high amount of cognitive effort?
Research on the effort-accuracy framework yields extensive evidence that decisions
that call for costly effort induce the decision maker to apply a decision strategy that
15
balances effort and attainable accuracy (Iglesias-Parro et al., 2001; Payne & Bettman,
1992). While these findings apply to decisions where effort and accuracy coincide in
the present (Soman, 2004), the framework remains silent on how these effort
investments impact subsequent decisions. Accordingly, it seems promising to extend
the insights from effort and accuracy. An approach that may be appropriate for
explaining how customers track cognitive investments to forthcoming decisions is the
concept of switching costs. This concept and its implications for the present research
are reviewed in the following section.
2.2.2 Switching Behavior and Cognitive Lock-in
Consumer switching costs represent the "perceived economic and psychological costs
associated with changing from one alternative to another" (Jones et al., 2002, p. 441).
Contrary to its primary conceptualization, recent research argues that switching costs
are multi-dimensional and encompass three different types of costs. More precisely,
switching costs can be broadly differentiated into continuity, learning and sunk costs
(Jones et al., 2002; Patterson & Smith, 2003). It appears that the first two categories
are especially important in a long-term perspective since they involve learning and
familiarization processes (Jones et al., 2002). The latter, however, may also be relevant
for decisions that do not involve extended learning processes. As such, sunk costs
might provide an answer to the question on how already invested cognitive effort
affects subsequent decision steps. In general, the sunk costs phenomenon describes the
enhanced commitment of a decision maker to a prior non-recoupable investment
because of its psychological importance and received considerable support in
management and consumer decision making research (see e.g., Arkes & Blumer, 1985;
Gourville & Soman, 1998; Jones et al., 2002). A large part of the research on sunk
costs is limited to economic costs, showing that past monetary investments can hardly
be ignored or written off by the decision maker, because doing so appears wasteful and
barely justifiable (Arkes & Blumer, 1985; Staw, 1976). This effect is also evident for
monetary equivalent investments of time (Soman, 2001 b ).
Apart from factors such as temporal and financial costs, researchers agree that
cognitive effort also raises switching barriers (Burnham et al., 2003; Fornell, 1992;
Soman, 2001b). To date, however, only anecdotal evidence and initial empirical
research provides support for this notion. That is, studies in the field of repeated
consumption revealed that consumers become locked-in to particular products because
of cognitive effort savings over time of usage. For example, Johnson and colleagues
(2003) found that efficiency gains in website usage make it costly for the user to
16
switch to another website. In their study, internet users became cognitively locked-in
to the particular website due to having already invested cognitive effort. Motivated by
this finding, Murray and Haubl (2007) investigated the phenomenon of cognitive lock-
in in the domain of tangible product consumption and a controlled laboratory setting.
They showed that the amount of experience and so the amount of effort reduction
increased incumbent product preferences. Similarly, in a series of three experimental
studies, Zauberman (2003) examined the relevance of information cost structures on
switching rates and lock-in. According to his results, cognitive lock-in in the
environment of internet usage is induced by initial information costs (i.e., the effort it
takes to process and finalize a search task) that reduce additional search and switching
behavior of the user. These findings imply that, over time, cognitive costs associated
with an initial product usage increase the probability that the buyer becomes
cognitively locked-in and therefore remains with the incumbent product (Johnson et
al., 2003; Murray & Haubl, 2007; Zauberman, 2003). Although these results fall into
the category of learning costs, they provide initial evidence for the lock-in impact of
cognitive investments. This gives rise to expect that a cognitive lock-in also occurs
when the decision maker solely arrived at a decision without prior payment or usage of
the product purely because of the exertion of cognitive effort.
This assumption is underpinned by a study recently conducted by Cunha and
Caldieraro (2009). In two experiments the authors provide initial empirical support
that also behavioral investment, such as cognitive effort expenditures, influence future
choices. In both studies, participants were likely to fall prey to behavioral sunk costs
that hindered future switching decisions. In particular, participants were less likely to
switch to a new product when their initial product selection required higher cognitive
effort investments. The authors reasoned that initial behavioral investments triggered
justification processes (cf. Kim & Labroo, 2011 for a similar argument). Although
research on the behavioral sunk-costs effect is still in its infancy (Cunha & Caldieraro,
201 O; Otto, 2010) it provides important preliminary insights for the current research.
One can expect that effortful initial reservation decisions raise switching barriers. High
cognitive effort investments may therefore create a cognitive lock-in situation that
prevents the customer from switching to the upsell offer. This cognitive lock-in
suggests to the decision maker that it might not be justifiable to switch to the upsell
offer, because otherwise initial investments on cognitive effort become sunk.
17
To sum up, research findings on the effort-accuracy framework and cognitive
switching costs allow for two conclusions: First, before and during information
acquisition and information processing a decision maker (i.e., the customer) strives
minimizing the amount of exerted cognitive effort. Second, when the decision maker is
forced to expand costly effort into a decision task due to the complexity of the decision
problem he becomes trapped by a cognitive lock-in. This lock-in situation causes the
decision maker to stick with an effortful decision. This implies that he or she remains
with his or her initial decision even in the light of a superior choice option. Otherwise,
the invested cognitive effort becomes sunk. Once exerted, cognitive effort acts as a
source of justification and prevents the customer from switching. In this regard, recall
the introductory example of the hotel reservation. If you invested much time and
cognitive strain into your decision on which hotel room to reserve, you might become
caught up in a cognitive lock-in, "forcing" you to stick with your initial reservation
decision even in the light of a superior room option.
Therefore, in order to realize the upsell one question remains: How can the salesperson
present the upsell arguments so as to release the customer from the cognitive lock-in?
In other words, which arguments are best suited to provide a stronger source of
justification relative to the exerted cognitive effort? In order to answer this question,
the following chapter will examine the potential impact of the message's frame.
2.3 The Upsell Offer: The Moderating Role of the Message's Goal Frame
Research on human decision making demonstrated that people are sensitive to the way
in which problems and decision options are presented (Shafir, 1993; Tversky &
Kahneman, 1986). That means decision makers are susceptible to the "frame", that is
the specific structure, of a message. Beneath a whole set of framing techniques,
message framing (goal framing in particular) presents a commonly encountered
communication approach in marketing and especially in service communications
(Putrevu, 2010). Hence, determining the relative persuasiveness of different frames in
upsell communications seems essential in the current context. The next two sections
will first give a short introduction to message framing, goal framing in particular, and
review significant results on its effectiveness. As a second step, goal framing will be
applied to the context of upselling.
18
2.3.1 Introduction to Message Framing
In order to illustrate how framing can be used to release the customer from a cognitive
lock-in, it is first necessary to gain a more pivotal understanding of how messages can
be framed and which technique is most appropriate in the current context. The
expression of message framing covers a diverse set of framing techniques that have
gained heterogeneous attention in research so far (O'Keefe & Jensen, 2006). These
framing techniques share a common feature: they provide information either in
positive or in negative terms (Putrevu, 2010). In their seminal work on framing effects,
Levin and colleagues (1998) differentiate between three different framing
manipulations according to the underlying mechanisms and consequences. Their
taxonomy encompasses risky choice framing, attribute framing, and goal framing.
Risky choice framing was first introduced by Tversky and Kahneman (1981) and
builds upon prospect theory (Kahneman & Tversky, 1979). According to this theory,
risky choice options are more attractive in light of a negative message frame. Contrary,
decision makers favor a low-risk choice option when it is presented with a positive
frame (Smith & Petty, 1996; Tversky & Kahneman, 1981). Consequently, framing the
outcome of a risky choice as either positive or negative evokes a preference reversal.
Notably, this kind of framing is limited to risky choice situations, like the well
documented "Asian disease problem" (Tversky & Kahneman, 1981 ). Due to this
limitation, risky choice framing seems not to be applicable to the current service
context.
In contrast, attribute framing, as a second framing technique, is equally applicable to
risky as well as non-risky situations (Levin et al., 1998). A single attribute or
characteristic of an object or event is subject to a positive or negative frame (Levin et
al., 1998; Putrevu, 2010). A prominent example includes framing beef as either "75%
lean" or "25% fat" (Levin & Gaeth, 1988). In general, due to a positive cue effect,
decision makers favor the positive rather than the negative attribute frame (Levin et
al., 1998; Putrevu, 2010; Smith & Petty, 1996). In contrast to risky choice framing, the
main dependent variable in attribute framing concerns an evaluation of the object or
event rather than a choice between two independent options (Levin et al., 1998).
Because the present research is concerned with whether the customer chooses the
upsell or not, attribute framing may not contribute to the research question.
The third framing alternative, goal framing, that received much attention in research
on persuasive consumer communication seems most qualified to meet these
requirements. According to goal framing, messages can be either framed in terms of
19
gams or losses. 7 Gain- or positively framed messages emphasize the advantages
associated with compliance, i.e., the advantages of adopting the communicator's
request. However, loss- or negatively framed messages emphasize the losses and
disadvantages linked with noncompliance, i.e., the disadvantages of failing to comply
with the communicator's suggested course of action (Levin et al., 1998; O'Keefe &
Jensen, 2006). In contrast to other framing techniques, goal framing manipulations
promote the same behavior, by framing the relationship between the decision maker's
behavior and goal attainment either in a positive or in a negative way (Putrevu, 2010).
Although most of the studies on goal framing can be found in the field of health-
related persuasion (Levin et al., 1998; Meyerowitz & Chaiken, 1987; Rothman et al.,
1993; Salovey et al., 2002), recent publications also apply goal frames to consumer
choice settings (Chang, 2008; Ganzach & Karsahi, 1995; Homer & Yoon, 1992;
Putrevu, 2010; Putrevu & Lord, 1994; Roggeveen et al., 2006; Yi & Baumgartner,
2008). A well-known example of goal framing in the domain of consumer choice has
been documented by Putrevu (2010). He explored consumer responses towards goal-
framed ads for an airline. The gain-framed advertisement for the airline emphasized
the on-time arrivals associated with choosing this carrier. Accordingly, the loss-framed
advertisement stressed the time delays and missed connection due to not flying with
this airline. As a result, flying with the promoted airline was beneficial in both
conditions. This logic is also well applicable to the current context of upsell offers,
because (a) choosing a travel service is commonly a low risk decision and (b)
upselling is all about promoting the same behavior, i.e., the acceptance of the upsell
offer. Consequently, the following section will provide a goal frame view of upsell
offers and present a so far neglected reasoning on the underlying mechanism.
2.3.2 A Goal Frame View of Upsell Offers
Due to the fact that an upsell embodies a superior product or service option, the upsell
message can either highlight the advantages of adopting the upsell offer or the
disadvantages of remaining with the initial reservation decision. Drawing on the hotel
example from the introductory section, a positively framed upsell offer highlights
gains and advantages of the superior room so as to induce the goal means to ensure
positive outcomes (e.g., "When you decide upon this room [ upsell] you may have a
relaxed stay due the comfortable room size."). Alternatively, a negatively framed
7 When referring to different goal frames, some authors rely upon the "gain/loss" distinction (c£ O'Keefe & Jensen, 2006)
while other authors use the ''negative/positive" labels (cf. Meyers-Levy & Maheswaran, 2004). However, both
classifications can be used interchangeably (Levin et al., 1998). To refer to the valence of the message frame, the present
dissertation will primarily apply the "gain/loss" distinction.
20
upsell offer would stress the potential losses and disadvantages of maintaining the
initial reservation decision thus highlighting the avoidance of negative outcomes (e.g.,
"When you stick to your reservation, your stay might not be as comfortable as it could
be due to the small room size."). Both messages promote the same offer - the upsell
offer. But which frame is more persuasive and accomplishes to release the customer
from the cognitive lock-in of the first decision stage?
In order to answer this question, assessing the relative persuasiveness of loss- and
gain-framed offers is key. Recent meta-analytic reviews showed mixed evidence for
the effectiveness of both types of goal frames (Levin et al., 1998; O'Keefe & Jensen,
2006, 2008). This heterogeneity may be due to various competing mechanisms that
have been suggested to account for the differential persuasive strength of loss and gain
frames (Levin et al., 1998; Smith & Petty, 1996). Furthermore, procedural and
contextual variations between studies (e.g., health vs. consumption) hamper the
transferability of results (Chang, 2008; Krishnamurthy et al., 2001). A vast body of
research suggests that loss-oriented messages have a stronger persuasive influence on
choice than gain-framed messages (e.g., Ganzach & Karsahi, 1995; Homer & Yoon,
1992; Krishnamurthy et al., 2001; Levin et al., 1998; Meyerowitz & Chaiken, 1987).
This result is often attributed to the well documented negativity bias that describes a
decision maker's inclination to be more sensitive towards negative information as
opposed to positive information (Herr et al., 1991; Krishnamurthy et al., 2001;
Maheswaran & Meyers-Levy, 1990; for a review, see O'Keefe & Jensen, 2008). In
terms of the negativity bias, negative information is considered more diagnostic than
positive information (Ahluwalia et al., 2000).
Although some authors qualify the negativity bias as robust (e.g., Cacioppo &
Gardner, 1999), various interactions between goal frames and context variables such
as price (Grewal et al., 1994), involvement (Maheswaran & Meyers-Levy, 1990),
gender (Rothman et al., 1993), need for cognition (Wegener et al., 1994), motivation
(Wilson et al., 1990) and processing style (Igou & Bless, 2007) as well as opposite
results have been documented (for a review see Krishnamurthy et al., 2001; Levin et
al., 1998). While many health-related studies suggest that a negativity bias exists (e .. g,
Meyerowitz & Chaiken, 1987), other studies in the domain of health and consumer
decision revealed a positivity bias (e.g., Chang, 2008) or failed to find any divergent
effects between loss and gain frames (Krishnamurthy et al., 2001; study 2).
21
Due to this inconsistency, a second process explanation that was developed in recent
years might provide a better foundation for framing effects and for the current research
context. This explanation considers anticipated emotions to act as potential mediators
of framing effects (Salovey et al., 2002) and therefore answers the call for a stronger
attention to affective as opposed to cognitive processes in framing research (Chang,
2008; Putrevu, 2010; Salovey et al., 2002). Although this explanation is still in its
infancy (Salovey et al., 2002), a few empirical results underpin its effectiveness.
Research on comparative ads shows that negatively framed ads encourage the
consumer "to think about potential losses they will incur from using the competitor's
brand" (Roggeveen et al., 2006, p. 115). In comparison, gain-framed ads prompt the
consumer to think about potential gains that might be attached to the superior brand
(Roggeveen et al., 2006). Accordingly, goal frames prompt the decision maker to
imagine the consequences and attached feelings that come along with their decisions.
While positive frames are found to evoke more positive affect (Chang, 2008), it shows
that loss-framed messages are akin to fear-inducing appeals (O'Keefe & Jensen,
2008). It appears that negatively framed messages are comprised of two components:
One that induces fear or related negative emotions by portraying the impending threat
and one that presents a recommendation on how to escape (Gierl et al., 2000; O'Keefe
& Jensen, 2008). This notion is in line with Yi and Baumgartner (2008) who showed
that messages stressing either gains or losses facilitate different kinds of imagery and
thus the anticipation of distinct future feelings. These anticipated emotions describe
"the feelings a person simulates when considering performing (or being blocked from
performing) the target behavior" (Salovey et al., 2002, p. 402). In fact, studies in the
field of consumer decision showed that anticipated emotions, such as pleasure and
pain, guide the decision maker's behavior and thus the decision outcome (Mellers &
McGraw, 2001).
It seems likely that these results can also be extended to the current context. Similarly
to the fear explanation, it can be proposed that negatively framed upsell messages
introduce different levels of anticipated regret, that is the "negative, cognitively based
emotion" (Zeelenberg, 1999, p. 94) decision makers anticipate if they had decided
differently (Connolly & Zeelenberg, 2002). In a similar vein, Janis and Mann (1977)
propose that the salience of relative losses evokes anticipated regret. According to the
above cited studies, loss-framed upsell messages that focus on the inferiority of the
initial choice would prompt the customer to think about potential disadvantages that
might arise when he or she remains with the initially chosen product or service option
(Roggeveen et al., 2006). Consequently, loss-framed upsell messages would encourage
22
the decision maker to anticipate the negative feelings associated with the decision to
stick to the initial reservation. In other words, loss-framed upsell offers may encourage
the customer to anticipate the regret that would result from inaction, i.e., anticipated
inaction regret (Sevdalis et al., 2006). In contrast, gain-framed upsell messages would
enable the customer to imagine the potential advantages of the upsell offer and may
not encourage the decision maker to anticipate any feelings of regret. Recent studies
on regret underpin these assumptions, by demonstrating a connection between
negative and positive outcomes and the short-term feeling of regret (Gilovich &
Medvec, 1994; Landman, 1987).
The following example will easily exemplify these arguments. Therefore, recall the
introductory scenario of the hotel reservation. Imagine you arrive at the hotel and the
attentive desk clerk tells you, considering the many bags you are carrying with you,
that the room you had initially reserved might become too small, leaving too little
space for all your belongings. Instead, she offers you a more spacious room for 10
Euros extra charge. In that moment you start to imagine how you would feel when the
desk clerk was right, i.e., when you enter the reserved room and note the lack of space.
You would probability regret your decision to stay with your reservation.
Consequently, in the service encounter you start to anticipate the regret that would be
tied to a wrong decision. In that moment, the negative arguments of the desk clerk may
provide a source of justification for you to switch to the upsell offer.
Now imagine a somewhat different example. The desk clerk tells you about a special
offer and describes the superior room in terms of the associated advantages. For a
premium of 10 Euros you can switch to a spacious room with more amenities instead
of the room you had initially reserved. Again, in that moment you start to imagine
your feelings when the desk clerk was right, i.e., when you enter the superior room and
note the ample space. This might feel positive. However, there is no clue in the
message that induces negative feelings on your reservation decision. Continuing with
your initial reservation would scarcely induce any feelings of regret. Compared to the
example above, the question asks if these arguments are strong enough to persuade the
customer and to deliver argumentative strength. The answer may depend upon the
initial choice step and thus on the amount of cognitive effort invested in this first
decision. Cognitively effortful decisions, as specified in the former sections, may raise
switching barriers and create a lock-in situation. Consequently, only strong arguments
that counterbalance these switching barriers may persuade the customer of the
favorable upsell offer. In terms of goal framing, loss appeals seem to unfold their
persuasive power since they enhance the anticipation of inaction regret. Compared to
23
gain-framed upsell offers they may be strong enough to cancel out already invested
cognitive effort. Different outcomes are to be expected when cognitive effort was low
and no or little lock-in occurred. When weak switching barriers exist, comparatively
weaker arguments may suffice to persuade the customer of the upsell offer.
These arguments suggest that the goal frame of the upsell offer moderates the
influence of cognitive effort on the customer's final decision. Consequently, the final
decision of the customer is a trade-off between the cognitive effort invested at the
initial choice stage and the argumentative strength of the upsell offer. One can assume
that the customer finally decides in favor of the option that is most justifiable and
therefore based on the best overall reason. It seems likely that cognitive effort and goal
framing trigger justification processes through the induced anticipated inaction regret.
The next paragraph will present a justification-based explanation for the interplay
between both independent variables and accounts for the key mediating role of
anticipated inaction regret and decision justifiability.
2.4 The Final Decision: A Question of Decision Justification
Generally speaking, decision makers favor choice options that are easy to justify and
that mitigate regret (Connolly & Zeelenberg, 2002). This assumption is a basic
premise in the research on reason-based choice (cf. Shafir et al., 1993) and decision
justification synopsized in the decision justification theory (DJT) (Connolly &
Zeelenberg, 2002). This theory "is concerned with how well-justified decision makers
themselves perceive a decision to be, the justification they themselves consider
reasonable for the way it was made, and the self-blame and regret they experience
when they do not see the decision as justifiable" (Reh & Connolly, 2010, p. 1407).8
Apparently, decision justification is tightly connected to the experience or anticipation
of regret, that is, the "negative, cognitively based emotion" (Zeelenberg, 1999, p. 94)
decision makers experience or anticipate if they had decided differently (Connolly &
Zeelenberg, 2002). It shows that decision justification theory accounts for many
contradictory findings of former studies in decision research (for a review see
Connolly & Zeelenberg, 2002) and weakens the so far robust link between action and
regret (Connolly & Reh, 2003). In contrast to norm theory (Kahneman & Miller, 1986)
and the well documented status quo bias (Ritov & Baron, 1990), research on decision
8 A related, but conceptually distinct, understanding of justification is described by research on accountability (Lerner &
Tetlock, 1999; Tetlock, 1992). A decision maker who feels accountable is concerned about meeting external standards,
implicitly or explicitly set by others. While internal and external expectations may possibly coincide, there is no necessary
cause to that (Reb & Connolly, 2010).
24
justification provides evidence that inaction is not preferred a priori but depends on a
person's assessment of the justifiability of an action as opposed to inaction (Reh &
Connolly, 2010). For example, in a series of experiments on vaccination decisions,
Connolly and Reh (2003) showed that the intention to vaccinate is a function of the
person's assessment of anticipated regret and decision justifiability. Furthermore, in
the context of consumer switching decisions, Inman and Zeelenberg (2002)
demonstrated that negative prior experiences with a product provide reasons to justify
a decision switch and heighten the anticipated regret associated with a repeat purchase
decision. These studies show that, because human decision makers are regret averse,
they try to regulate their decision by increasing decision justifiability and consequently
reducing experienced and anticipated regret (Reh, 2008; Simonson, 1992; Zeelenberg
& Pieters, 2007). This justification mechanism is well illustrated in a study conducted
by Simonson (1992). In one experiment students had been primed to imagine the
feeling of regret followed by a product choice between a higher-priced premium brand
and a cheaper, less known brand. Participants who were prompted to anticipate regret
were more likely to choose the familiar, high-quality brand instead of the unfamiliar
one. In line with decision justification theory, Connolly and Zeelenberg (2002)
interpret these results as follows: "thinking about regret led them [the participants] to
look for justifications for their choices, and the safer brand [ ... ] offered the
justification they were looking for" (p. 214). This means, that the study participants
chose those products that were supported by the strongest overall reasons in order to
mitigate regret and to remain justifiable (Shafir et al., 1993).
Now, assuming that negative frames evoke stronger anticipated feelings of inaction
regret than gain frames, these findings would suggest that a loss-framed upsell offer
triggers justification processes. In the light of possible losses it seems unjustifiable to
stick with the initial reservation decision, even if this decision was cognitively
effortful. It can be concluded, that loss-framed upsell offers may cancel out already
invested cognitive effort while providing strong arguments for a decision switch. In
comparison, gain-framed upsell offers highlight the advantages of upsell offers and
thus may not induce strong anticipations of regret that might result from inaction.
Quite the contrary, in the high effort condition switching to the gain-framed upsell
offer would induce regret since the invested cognitive effort would be sunk without a
strong justification. While effortful initial decisions may call for a strong justification
for a decision shift towards the upsell offer, less effortful reservation decisions may
not. Consequently, in the low effort condition gain-framed upsell offers should also
provide sufficiently justifiable reasons to decide in favor of the upsell offer.
25
Note that these propositions assume that the customer himself exerted cognitive effort
at the initial choice stage and that the extent of this effort interacts with the message
frame. Another scenario that is quite usual in a service setting is that the initial
reservation decision is completely relinquished to a surrogate buyer who decides on
behalf of the customer. Just like the customer, the surrogate also has to invest a certain
amount of cognitive effort in the reservation decision. The following chapter will
revisit the initial decision step by incorporating a surrogate shopper.
2.5 Initial Decision Revisited: The Impact of a Surrogate Shopper
In many services, especially in the travel industry, customers often delegate their
initial decision on which hotel room to book or car to rent to a third party that acts as a
surrogate shopper. The World Travel Monitor shows that, despite increased internet
usage, in 2011 up to 30% of all international travel bookings were performed through
surrogates such as travel agents (ITB, 2011 ). Thus, it seems important to investigate
the influence of a surrogate's reservation decision on the decision maker's final
choice. In what follows, the next paragraph will shortly review the concept of
surrogacy and will provide a clear conceptualization of surrogate shoppers as used in
this dissertation. The second paragraph will focus on the concept of the surrogate's
cognitive effort, i.e., perceived effort, as an equivalent to the cognitive effort of the
customer at the first decision step.
2.5.1 Surrogates in Services
A surrogate shopper (or surrogate buyer) can be defined as "a commercial enterprise,
consciously engaged and paid by the consumer [ ... ] to make or facilitate selection
decisions on behalf of that consumer" (Hollander & Rassuli, 1999, p. 102f.).
Surrogates build their decisions and preferences on a multitude of information sources
and therefore differ from store-employed salespeople who do not recommend
competing products or services (Hollander & Rassuli, 1999). As opposed to opinion
leaders and family members who perform surrogate tasks sporadically, surrogate
shoppers are specialized in a certain product or service category and perform some or
all of the decision making tasks for payment in return (Aggarwal & Mazumdar, 2008;
Hollander & Rassuli, 1999). Travel agents, architects or financial advisors are typical
examples of surrogate shoppers (Hollander & Rassuli, 1999). They act as human
decision aids by providing an institutional structure and expert knowledge to the
customer. Travel agents, in particular, manage to structure diverse travel opportunities
26
hence preventing the customer from information overload and providing support when
time is scarce (Bechwati & Xia, 2003; Hollander & Rassuli, 1999). Travel agents, just
as other types of surrogates, can perform some or all of the decision making tasks.
Thus, decision delegation encompasses three different components: attribute set
delegation, choice set delegation and choice delegation (Aggarwal & Mazumdar,
2008). A customer who retains the surrogate to identify the salient attributes that have
to be considered in decision making delegates the attribute set. Considering the initial
scenario, this means that the customer would assign a travel agent to identify the
characteristics of hotel rooms that are important to come to a reservation decision. In
addition, a customer who delegates the choice set requests the surrogate to narrow the
choice set on his behalf. Accordingly, this would mean that the travel agent singles out
which hotels or rooms are most suited to cover the customers' needs. Finally, the
customer can also delegate the (complete) choice to the surrogate, meaning that the
travel agent would decide on which hotel room to reserve. It is possible that the
customer delegates all three decision components to the surrogate but that might not
necessarily be the case. The extent of delegation is, besides other factors, a function of
the customer's expertise and the frequency of related decision situations (Aggarwal &
Mazumdar, 2008). It shows that infrequent or one-time decision situations especially
facilitate attribute and choice set delegation (Aggarwal & Mazumdar, 2008).
Moreover, a difference in expertise between the customer and the surrogate favors
decision delegation on all three manifestations (Aggarwal, 1998; Aggarwal &
Mazumdar, 2008).
The latter findings are of special interest in the present context, because it appears
obvious that booking decisions for hotel rooms, flights or rental cars are commonly
performed infrequently. In order to account for this situation, the current research
conceptualizes decision delegation as choice delegation that encompasses attribute and
choice set evaluation as well as the reservation choice itself. Consequently, one can
assume that when a customer delegates the complete reservation decision to a
surrogate, the customer is likely to become locked-in to this reservation because he or
she neither knows the choice set nor the associated attributes and thus depends upon
the surrogate's expertise (Jaakkola, 2007).9 The strength of this lock-in, however, may
depend upon the effort the surrogate invested in the reservation decision. The next
section will elaborate on the key role of an employee's effort.
9 The present dissertation assumes that the surrogate sells his best advice to the customer (Hollander & Rassuli, 1999). This
implies that no undisclosed conflicts of interest should occur. Thus, the conceptualization of surrogates' decisions excludes
agency related problems (for further remarks on agency problems in marketing see e.g., Bergen et al., 1992).
27
2.5.2 The Role of Perceived Effort
Similar to the customer, the surrogate shopper also has to exert cognitive effort to
come to a reservation decision at the initial choice stage of the upselling process. The
exerted effort acts as a clue for the customer to evaluate the surrogate's decision work
(Bechwati & Xia, 2003; Mohr & Bitner, 1995). Contrary to the customer's effort, the
surrogate's effort has to be observable and thus assessable for the customer. This is
why the surrogate's or service employee's effort is termed perceived employee effort
(Huang, 2010) or perceived effort (Mohr & Bitner, 1995). Perceived effort refers to
"the amount of energy a customer believes an employee has invested on their behalf'
(Huang, 2010, p. 195). Scholars suggest that unlike one's own cognitive effort,
consumers welcome effort expended by others on their behalf (Bechwati & Xia, 2003;
Kahn & Baron, 1995). In other words, because customers are cognitive misers that
seek accuracy (see effort-accuracy framework Chapter 2.2.1), they appreciate effortful
decision processes of surrogate shoppers. Several experiments conducted by Kahn and
Baron (1995) underpin this paradox. Confronted with different decision scenarios,
study participants were prompted to choose the decision rule they were likely to apply.
Whereas most of the participants chose the effortless one-factor rule for themselves
they wanted their agents to apply more complex and effortful non-compensatory
decision rules (Kahn & Baron, 1995). For example, in a medical scenario on treatment
selection only fifteen percent of the participants felt comfortable using an effortful
decision rule. Surprisingly, twice as many subjects indicated that the physician should
apply this decision rule instead (Kahn & Baron, 1995; study 3a). In addition, in a study
comparing human and electronic decision aids, Bechwati and Xia (2003) illustrate that
unlike electronic decision aids (i.e., computers), consumers want their human decision
support (e.g., a career advisor) to exert cognitive effort and thus act as an effort saver.
Extending these findings, some studies have addressed the question of how the effort
exerted by others (vs. one's own effort) affects consumers' evaluation of the decision
process and the service encounter (Bitner et al., 1990; Folkes, 1984; Huang, 2008,
2010; Keaveney, 1995; Specht et al., 2007). These studies show that customers take
heed of an employee's effort exerted in a service context and that these perceptions
influence the degree of satisfaction with the service. Using critical incident technique,
Bitner and colleagues (1990) and Keaveney (1995) found that, when asked for highly
satisfying or dissatisfying incidents, customers often refer to the amount of effort
invested by the service employee. Moreover, in a study on service failures, Huang
(2010) showed that perceived employee effort at solving a service problem affects the
level of satisfaction, repurchase intentions, as well as word-of-mouth.
28
Although, however, the last three studies did not differentiate between employee's
cognitive effort and physical effort, they allow for two conclusions that matter in the
current context. First, although decision makers are cognitive misers they expect a
surrogate shopper to exert much cognitive effort on their behalf. Second, perceived
employee effort counts, meaning that surrogates' decisions that involved much
(cognitive) effort are more relevant to the customer than effortless decisions of others.
Consequently, it can be assumed that the amount of perceived effort reinforces the
lock-in of the customer that accrued due to choice delegation. It seems likely that the
customer is stronger bound to the surrogate's decision when the surrogate invested a
higher amount of cognitive effort as compared to lower effort investments. The
complete argumentation can be easily illustrated using the introductory example of this
dissertation.
Imagine you plan your next vacation and need a hotel room. Due to some time
constraints you decide to retain a travel agent to evaluate and choose a hotel on your
behalf. Since you completely delegated the first decision step to the travel agent, you
neither know the product options nor the decision rule that was applied by the travel
agent. Consequently, assuming that the surrogate appears competent, decision
delegation may build up switching barriers. It may seem hardly justifiable for you to
switch to an upsell option the desk clerk might offer you when you arrive in the hotel.
Hence, it seems likely that decision delegation at the initial decision step leads to a
lock-in situation for the customer.
Additionally, imagine the travel agent exerted much cognitive effort in the reservation
decision. Because customers appreciate effortful decisions of agents, this reservation
would provide a positive experience. In general, positive prior experiences raise
switching barriers and thus provide the customer with strong reasons to remain with
the reservation (Inman & Zeelenberg, 2002). 10 Hence, when the customer observes the
surrogate to exert much effort at the initial decision step, continuing with this decision
seems to be more justified than switching to the upsell offer. Additionally, if the
customer relinquishes the entire decision control to a surrogate buyer, the service
outcome becomes unknown and ambiguous to the customer boosting the importance
of perceived effort as a service clue (Berry et al., 2006; Mohr & Bitner, 1995). It can
be therefore predicted that delegating the reservation decision on the first choice stage
to a surrogate buyer principally leads to a lock-in to this decision that is additionally
enforced when perceived effort is high. Hence the probability that the customer
10This rings true when the employee is perceived as competent and efficient (Mohr & Bitner, 1995). Otherwise, when the
employee has to invest cognitive resources due to a lack of ability, this argumentation should cease to hold.
29
chooses the upsell offer is low. In order to convince the customer to switch to the
superior upsell offer, the service employee at the second choice stage needs strong
arguments. As shown before, negative frames bring forth strong reasons for a decision
switch by highlighting the negative consequences of non-acceptance compared to
positively framed upsell messages that highlight the advantages of accepting the upsell
offer. According to Montgomery (1983) and Inman and Zeelenberg (2002), decision
makers decide in favor of the decision option that is associated with the greatest
number of supporting reasons. If the surrogate invested much cognitive effort at the
initial choice stage, the customer has two compelling reasons to continue with the
reservation decision: First, the customer does not know the choice set but expects the
surrogate to sell his best advice and, second, effortful decisions of the surrogate
provide positive experiences. It can be assumed that these two arguments loom larger
than the arguments provided by loss- and gain-framed upsell offers. Considering high
perceived effort, a decision switch might be neither justifiable in light of a negative
frame nor in light of a positively framed upsell offer.
Finally, imagine the travel agent "only" invested a low amount of cognitive effort in
the reservation decision. In this case, perceived effort would not provide any
arguments to stick with the reservation decision. But again, the customer does not
know the initial choice set. Thus, in order to convince the customer of the upsell
option, the desk clerk at the hotel would have to provide arguments that balance the
knowledge gap. As opposed to gain-framed messages, negatively framed upsell offers
seem likely to meet this requirement since they highlight the disadvantages of the
reserved room option and implicitly provide information on the upsell option.
Accordingly, they may provide stronger reasons for a decision switch.
Summarizing, this section revisited the initial choice step and modified decision
responsibility for the initial reservation decision. It shows that, unlike one's own
reservation decision, decision makers expect their surrogates to exert extensive
cognitive effort. Moreover, it becomes apparent that choice delegation at the initial
choice step generates a knowledge gap between the service employee and the
customer. These theoretical assumptions lead to quite different consequences for the
final choice of the customer in the service encounter. To include these arguments in
the overall context of this dissertation, the next section will provide a unifying
conceptual model that will link the different decision steps and associated decision
components with each other.
30
2.6 Linking the Steps: The Conceptual Model
The arguments that have been developed so far are summarized in the conceptual
model in Figure 2-2. The model appears as a procedural combination of those
variables that are supposed to influence the customer's final decision in terms of
choice probability and perceived expensiveness of the upsell price. As such, the model
seeks to explain why and how the customer's final choice is influenced by the initial
reservation decision and the presentation of the upsell offer. More precisely, the model
accounts for the obstructive effect of initial cognitive effort investments either exerted
by the customer or a surrogate shopper and the moderating impact of the upsell offer's
goal frame. As such, different goal frames are expected to unfold their persuasive
appeal contingent upon the amount of initial effort investments.
Moreover, the model postulates that the final choice of the customer is a question of
decision justification. While the initial effort investment is expected to establish
arguments against a decision switch, different message frames are likely to offer
varyingly strong arguments in favor of the superior offer. Eventually, the customer is
supposed to decide in favor of the option that mitigates anticipated inaction regret and
is thus more justifiable.
Because the model is primarily designed to apply for services especially in the travel
industry, it distinguishes between the following settings. First, while the initial
reservation decision can be made using both electronic reservation systems and human
decision aids, the upsell offer as well as the final decision is assumed to occur in the
dyadic interaction between the service provider and the customer - the service
encounter. Second, in order to account for industry-specific characteristics at the initial
choice step, the model postulates that cognitive effort investments of customers and
surrogates differ in their impact on the customer's final choice.
The following chapter will convert and substantiate the theoretical foundation and
conceptual model in concrete hypotheses that will be put to test in the empirical part of
this dissertation. Most of these hypotheses will be proposed for the upper conceptual
path, that is, the joint impact of the customer's initial cognitive effort and the
message's goal frame on the final choice of the customer. The lower conceptual path
that illustrates the influence of a surrogate shopper, however, will extend these
findings and will provide initial evidence for future research attempts.
11.u.st.urru.u".ts Ifiitbtl
lleciiio.n
lllllllllllllllllllllllllllllllltllllllllllllll
11111111111111111111111111111111111111111111111
Surrogate's. Initial
Decision
11111111111111111111•11111111111111111111111
1111111111111111111111111111111111111111111111
1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111
Decision
Justifiability
&
Anticipated
Inaction Regret -
1111111111111111111111111111111111111111~1111111111111111111111111111111111111111
:::::::::::::::::::::::::11111~1111~~:::::::::::::::::::::::::
~
~· :::
~
t ..
g
~
~
~
~
~ -
~
~ - -..·vd1 ;i~ .:s h:~:rd.ry·.:-; 1fl t:S~ btl~i}:-t~~}HJ y~~l~ t-'-:in J1r1.d :1 rni;_;fr.,rt>'i.t .\.b.Jf\~·.;::r Fti-; hr~t1k.tn~:t )'\)1i ..::.;;.;.1~ :,l:irr...-~~
)'O•,~r~ .. •;·lf tH ~ht: h1~m.·! .:.•t u~,l::~ f~X· :.-U(•~j'l_ .:;,..:•J'\"'(~\:'
Tb0 n.:.:.1nJ.: -~a.,::urrip::::;:>('~ 2;;. ~tl Ji'.l. lt h({" t\'\;(1 t:~.i..h >'!:Ud ~'i1{1:"S \ViYi. Add.Vi,nJ:l:ty. 1J1>: r0<.>cH l~ ·i:-',~:.:~:>J.x:d ._y_~~h :i; ~t~.t\:'
l'Cl:·J3 .) iHit..ithR~·:-r:~ )'(t•.L(Xt• Sln...'i~~ en;i[L)'­
t~1J:: T"X'H~ i:t "~ll.•~:pN} w~J·h :.i ~1f('._... .>.\ ~rnnib:1:. :l '-Ni~{!~~ Jr;:;k ~1~ ~-~·-t:)! .:,{'.i il b-w,-! t:~~~ nx:ti:~ :.~~n k ~
rh::; f•)0W(·<:m:iJ:il~•.tl$Sl::!~, 1~i 'iZ! m H :1'-l::. !'/,;) b-ixt.;; *.Ni {l.~r~~·t~ \q'f:~ .. '\d\.~i1k1m-1:~~y, ~h{< :~i«m ;s Ni~·ipr•tX1 '>'.-i1h •i: ~,~,:)~_.
;) :n:n:tY.::r :.:>:\ \v..:-.ll ;i~ :.i h~5!·~~tY'i!L \L"!-n 1-1ku ;~;)\'~ ~::i:.:. w:;;k::'.>1:": ·:;~rvi.\:{'. ·~l )':':on• t~t:;p,)X.':[ l'i t:J~': '.t)::i:(hn':~~sn ~':~u ~"'.";~n
fi.rJ1J, on ;,;n"')l'-;11 ··~h .30). Moreover, the manipulation check for the message's goal
frame indicated two significant main effects for each of the framing manipulations.
First, participants who were confronted with a loss-framed upsell offer felt more
strongly that the message focused on negative aspects (MLossFrame_LossCond = 4.83) than
participants in the gain frame condition (MLossFrame_GainCond = 3.21, F(l, 99) = 32.68,p <
.001). Second, in the gain frame condition participants felt that the upsell offer
emphasized the positive consequences (MGainFrame_GainCond = 5.91) more strongly as
opposed to the loss frame condition (MGainFrame_LossCond = 4.90, F(l, 99) = 18.39, p <
.001). As expected, there was no significant interaction effect between cognitive effort
and the message's goal frame (ps > .40). Consequently, it can be reasoned that both
manipulations were successful (for all means see Table 4-3).
55
Coefound Checks. Finally, participants' responses to the confound checks revealed no
significant differences (ps > .13). The results of two univariate ANOVAs indicated
that each of the four scenario combinations was perceived as equally credible and
comprehensible. It therefore appears unlikely that these variables significantly
confounded participants' response behavior.
Covariates. The subsequent analysis showed that only persuasion knowledge was a
significant covariate. A univariate ANOV A revealed that the different conditions did
not differ on persuasion knowledge (p > .36). Hence, persuasion knowledge shows to
be independent of the treatment effect and can therefore be included as a covariate into
the following analyses (cf. Field, 2009).
4.2.4.2 Hypotheses Testing
Choice Probability. Hypotheses la and lb predict that the probability of choosing the
upsell offer would be more strongly affected by different goal frames when cognitive
effort expenditure on the initial choice step was high compared to when it was low. In
order to test these hypotheses, an analysis of covariance (ANCOV A) was conducted.
A 2 x 2 ANCOV A revealed a significant main effect for cognitive effort on choice
probability, providing initial support for the lock-in assumption (F(l, 98) = 4.67, p <
.05). More importantly, this main effect was qualified by a significant interaction
between cognitive effort and goal framing of the upsell offer (F(l, 98) = 4.44, p < .05).
This interaction is also depicted in Figure 4-4. The main effect for the message's goal
frame was not significant (F(l, 98) = 1.27,p > .26).
To follow up on the significant interaction effect, planned contrasts were performed
within both effort conditions. As Figure 4-4 indicates, when participants invested a
high amount of cognitive effort at the initial choice stage, upsell arguments with a loss
frame were more effective (Mioss = 3.38) than gain-framed messages (Mgain = 2.35,
t(49) = -2.28,p < .05). That is, participants were more likely to choose the upsell offer
and thus reject their initial reservation decision, when the upsell offer was framed
negatively as opposed to the positive frame. A quite different pattern of results was
found when the participants invested a low amount of effort in the initial choice.
Consistent with hypothesis 1 b, the probability of choosing the upsell did not vary as a
function of the message's goal frame within the low cognitive effort condition (Mioss =
3.55, Mgain = 3.80, t(50) = .46, p > .64). Summarizing, these results provide support for
hypotheses la and lb.
56
Perceived Expensiveness. Moreover, a second two-way ANCOV A was conducted in
order to test hypotheses 2a and 2b. The analysis revealed a significant main effect of
cognitive effort on perceived expensiveness of the upsell offer (F(l, 98) = 4.47, p <
.05) and an insignificant main effect of the message's goal frame (F(l, 98) = .31, p >
.58). More importantly, the analysis revealed a marginally significant interaction
between cognitive effort and message frame (F(l, 98) = 3.39, p < .07) (see Figure
4-4).
To gain a deeper understanding of the nature and direction of the interaction, two
planned contrasts were performed. It shows that participants in the high cognitive
effort condition perceived the upsell price as more expensive when they were
confronted with a gain-framed (Mgain = 5.13) upsell offer than when the upsell offer
was loss-framed (Mioss = 4.46, t(49) = 1.52, p = .068 (one-tailed)). Though the results
are only marginally significant, they clearly point into the hypothesized direction. In
contrast, when cognitive effort expenditure on the initial choice step was low the
message's frame did not have a differential impact on the perception of price
expensiveness (M1oss = 4.24, Mgain = 3.91, t(50) = - .71, p > .24 (one-tailed)).
Accordingly, hypotheses 2a and 2b were supported.
In order to show that perceived expensiveness is "conceptually reversed" to the
assessment of willingness to pay, in that higher ratings of expensiveness indicate a
lower willingness to pay for the upsell, an additional ANCOV A was conducted. This
analysis intends to legitimate the use of the perceived expensiveness measure as a
substitute for willingness to pay in the proceeding studies. The analysis revealed a
significant main effect of cognitive effort on willingness to pay (F(l, 98) = 6.64, p <
.05), an insignificant main effect of the message frame (F(l, 98) = .63, p > .43) and a
significant interaction between cognitive effort and the message's goal frame (F(l, 98)
= 3.93,p = .05). This interaction is also displayed in Figure 4-4.
Inversed to the perceived expensiveness measure, in the high cognitive effort condition
participants were willing to pay a higher amount of Euros for the upsell when the offer
was loss-framed (Mioss = 6.41) compared to when it was gain-framed (Mgain = 3.83,
t(49) = -1.70, p < .05 (one-tailed)). The framing of the upsell message had no
significant impact on participant's willingness to pay when cognitive effort was low
(Mioss = 7.62, Mgain = 8.57, t(50) = .60, p > .25 (one-tailed)). Note that the mean values
represent the amount of Euros. It shows that willingness to pay is conceptually
reversed to the employed perceived expensiveness measure. The results of all
ANCOV As and planned contrasts are illustrated in Table 4-2 and Table 4-3.
57
Table 4-2: Results of the ANCOVAs in Experiment 1
Dependent Variable F(l, 98) p
Persuasion Knowledge Choice Probability 4.04 p<.05
(Covariate) Perceived Expensiveness 3.68 p<.06
Willingness to Pay 5.64 p<.05
Cognitive Effort Choice Probability 4.67 p<.05
Perceived Expensiveness 4.47 p<.05
Willingness to Pay 6.64 p<.05
Message's Goal Frame Choice Probability 1.27 p>.26
Perceived Expensiveness .31 p>.58
Willingness to Pay .63 p>.43
Cognitive Effort x Choice Probability 4.44
p<.05
Message's Goal Frame Perceived Expensiveness 3.39 p<.01
Willingness to Pay 3.93 p=.05
Table 4-3: Mean Values for the Dependent Variables in Experiment 1
IDgh Cognitive Effort Low Cognitive Effort
Loss Frame Gain Frame Loss Frame Gain Frame
Choice Probability 3.38 (1.70) 2.35 (1.47) 3.55 (2.03) 3.80 (1.88)
Perceived Expensiveness 4.46 (1.64) 5.13 (1.46) 4.24 (1.46) 3.91 (1.91)
Willingness to Pay 6.41 (5.84) 3.83 (4.83) 7.62 (6.04) 8.57 (5.05)
MC Cognitive Effort 4.22 (1.13) 4.25 (1.20) 3.35 (1.17) 2.91 (1.06)
MC Gain Frame 4.64 (1.38) 5.54 (1.19) 5.16 (1.22) 6.28 ( .88)
MC Loss Frame 5.00 (1.34) 3.59 (1.49) 4.67 (1.46) 2.83 (1.47)
Note. The numbers in parentheses represent standard deviations. Willingness to pay was assessed in an open-
ended question, such that the numbers represent the amount of Euros per day a participant was willing to pay
additionally for the upsell offer. The remaining variables were assessed on seven-point scales, with higher
numbers representing higher mean ratings. MC is the abbreviation for manipulation check.
58
Figure 4-4: Cognitive Effort and Goal Framing Interactions (Experiment 1)
Choice Probability
4 "'""""""""""""""""""""""""""""""""""""""""""""""""""""'
I ::;; ___ ,w
3.5
~ ~''''"''''"''''''"'''"'~ 3.38
3 I
2.5 .,\. ..................................................................................................... ,
I 2.35
2 ·-,'"""""""""""""""""""""""""T""""""""""""""""""""""""";
Low Cognitive Effort High Cognitive Effort
Perceived Expensiveness
5.5 r 5:,:
5 ""'"""""""""""""""""""""""""""""""""'"
4.5 t . t ~,,,,,,,,,,,,,,,,,-$ 4 46
t ~~~~~ •
4 J"""""""""""""' """"""""""""""""""""""""""""""""""""
3.5
t 3.~J T •
Low Cognitive Effort High Cognitive Effort
Willingness to Pay*
9 ·'c""""""""""""""""""""""""""""""""""""""""""""""""""""""
8.57
8 """"""""""""""""" """"""""""""""""""""""""""""""""""""".
i 7.62~''"' : .._,,"'
7 .~ ........................................... ~':~'-"''''':;C .............................................. .
i .._,,,,,,,"' 6r-- ....... 6.41
: [ T 3:83 •
Low Cognitive Effort High Cognitive Effort
* assessed as additional Euros per night
......,.Gain Frame
%-''''''""Loss Frame
.,_.Gain Frame
%-""'""Loss Frame
~Gain Frame
59
4.2.5 Discussion
The first experiment tested the basic assumption that the amount of cognitive effort
invested at an initial choice stage determines the final choice of the customer and that
this influence is moderated by the goal frame of the upsell offer. More precisely, it
shows that customers are likely to become cognitively locked-in after exerting much
cognitive effort at the initial reservation task. However, framing the upsell offer in
terms of the disadvantages of the initial choice provides an effective tool to release the
customer from this lock-in. That is, in the high cognitive effort condition customers
were more likely to choose the upsell offer and to additionally pay for this offer when
the arguments were loss-framed as opposed to gain-framed arguments. When
confronted with loss-framed arguments participants rated the upsell price to be less
expensive than the upsell premium in the gain frame condition. It showed that gain-
framed arguments did not sufficiently demonstrate the value for the price premium as
compared to their loss-framed counterparts. As expected, when the cognitive effort
investment on the initial choice stage was low, participants were less bound to their
initial decision and thus more willing to accept the upsell offer. Consequently, framing
the upsell offer either in terms of losses or in terms of gains did not trigger differential
customer reaction in terms of choice probability, willingness to pay, and perceived
expensiveness.
Presumably, these results can be attributed to underlying justification processes. More
precisely, it can be assumed that effort investments raised switching barriers that
provided strong reasons to stick with an initial reservation decision. In order to
mitigate future regret participants needed strong reasons to justify a decision switch.
While loss-framed messages provided a strong source of justification, gain-framed
messages seemed to be less appealing. As a consequence, participants in the high
cognitive effort condition were more likely to choose the upsell offer when the upsell
arguments provided reasons to justify this decision switch. In order to directly test this
mediation process, study 2 will incorporate a number of mediating variables.
Although the results of the first experiment are promising, one may argue that the
results are limited due to the sample and industry selection. Although employing
student samples in the current context can be considered appropriate (see Chapter 4.1),
the next study will rely on a different sample as well as a different industry in order to
provide full support for the basic hypotheses 1 and 2.
60
4.3 Experiment 2: Cognitive Effort, Goal Framing and the Underlying
Process
The aim of experiment 2 was (a) to replicate the results of study 1 while incorporating
a number of important variations and (b) to provide a better understanding of the
process underlying the proposed effect. Hence, the following experiment tested the
hypotheses 1 (a, b ), 2 (a, b) as well as hypotheses 3 and 4.
4.3.1 Design, Participants, and Procedure
The experiment used a 2 (cognitive effort: high vs. low) x 2 (message's goal frame:
loss vs. gain) between-subjects design. The participants were 155 adults from a
professionally administered and paid German online panel (72 females, 83 males). On
average, respondents were 42.5 years old. In order to enhance situational involvement
for the studies topic, all subjects were screened for holding a driver's license before
starting with the experiment. Additionally, participants were randomly assigned to one
of the four conditions.
The procedure of experiment 2 was similar to the one outlined for the first experiment
except for some variations. That is, participants were asked to imagine they were
going on holiday and would like to reserve a car at a fictitious car rental firm. In the
first part of the experiment, participants were exposed to the reservation task that
manipulated the amount of cognitive effort. Given a rental period of one week and a
reservation price of 50 Euros per day, participants had to weigh four different car
options against each other in order to select one of them for reservation. The
manipulation of cognitive effort followed the procedure as outlined in the first study.
In contrast to the first study, the second part of the present experiment relied on films
rather than written scenarios to depict a service encounter and manipulate the
message's goal frame. Participants were asked to imagine they were the customer in
the service interaction that was shown in the video-vignette. Because the experiment
was online-based, an auditory check at the beginning of the experiment ensured that
every participant could clearly understand the spoken service interaction. Except for
the manipulation of the goal frame, both video vignettes were identical. Once the
video ended, the participants were asked to answer the dependent and process
measures and to fill out a series of manipulation and confound checks. Again, several
covariates were incorporated in order to control for individual differences between the
participants. Processing through the study took about 15 minutes on average.
61
4.3.2 Materials and Manipulations
In order to embed the manipulation of both independent variables - namely cognitive
effort and the message's goal frame - in one coherent and realistic story, all
participants were exposed to a scenario introduction. This introduction provided the
participants with information on their upcoming vacation and read as follows:
Soon it will happen - your next vacation is approaching! In this year's autumn
holidays of one week, you want to discover Germany and meet some friends on the
way. After beginning your road trip in Kiel, you want to rent a car to drive all the way
to Munich. You have already booked a flight to Kiel some time ago. So the only thing
you have to do is reserve a rental car, with which you can drive from Kiel to Munich.
You therefore visit the homepage of the car rental company "RentThisCar" and enter
the rental period of one week as well as your price limit of 50 Euros per day. Hence,
all you have to do is decide which car to reserve. The screens hot of the following page
depicts the options that are available to you.
Following, the participants were exposed to a screenshot of the homepage of a
fictitious company named "RentThisCar". This screenshot manipulated the amount of
cognitive effort that had to be invested in assessing the car options.
4.3.2.1 Manipulating Cognitive Effort
As in the first study, cognitive effort associated with decision making was manipulated
using different types of information representation. While enhanced cognitive effort
was manipulated by a list presentation in sentence format, lower effort was induced by
the alternative-by-attribute matrix. Due to the different industry, the nature of the
information that was provided in both conditions was different to the first study. That
is, the choice set encompassed four different car options each being described on six
different attributes. These attributes were retrieved from literature on car features
(Safer, 1998) as well as booking pages of established car rentals. Each automobile was
described in terms of number of doors, trunk capacity, hp (horsepower), transmission
type, gasoline consumption, and equipment. Hence, each car description comprised
reliability and luxury features as well (Safer, 1998). A number of pretests revealed that
both representation types required different amounts of cognitive effort. Hence,
cognitive effort was manipulated as depicted in Figure 4-5 and Figure 4-6.
62
Figure 4-5: Low Cognitive Effort Manipulation in Experiment 2
Figure 4-6: High Cognitive Effort Manipulation in Experiment 2
This cm features three doors The trunk has space for one la19e suncase Hie car offers 90 r:p arid has an
autamalic transmiss;an, '1he car is capable ni:19 Tile trunk !JBs :';pace for one
!a:ge and one small suitcase. Tne car o1lers 60 hp a:)d has an aut thrtle m:·ons aN:i tMe :run~ t;as space forone lerge and or:e smai! su•lcase. The •:aroifers 75 ~;p
and has a manual ;;aiwrnssi .40). Moreover, examination of both framing indices indicated that in
the loss frame condition, participants felt the upsell message conveyed more negative
aspects (MLossFrame_LossCondition = 5.12) as opposed to the message in the gain frame
condition (MLossFrame_GainCondition = 3.91, F(l, 151) = 20.50, p < .001). Participants also
felt that the upsell message delivered more positive aspects when it was gain-framed
(MGainFrame_GainCondition 5.63) than when the message was loss-framed
(MGainFrame_LossCondition = 5.14, F(l, 151) = 6.35, p < .05). Again, the results reveal that
69
the second independent variable as well as the interaction between the two
independent variables did not significantly affect both framing measures (ps > .18).
Consequently, the manipulation can be considered successful. Table 4-6 provides an
overview of the means of all manipulation checks.
Coefound Checks. Secondly, a number of confound checks were conducted to ensure
that the manipulation had only affected the intended variables. Therefore, three
univariate ANOV As were performed. The results revealed that the four conditions did
not differ in terms of credibility (p > .37), comprehensibility (p > .83) and perceived
realism (p > .66). These results provide evidence that all four conditions were equally
understandable.
Covariates. The following analysis revealed that gender was a significant covariate
(p < .05). The remaining possible covariates did not emerge as significant control
variables. These measures were therefore excluded from the following analysis.
4.3.4.2 Hypotheses Testing: Moderation
Choice Probability. It was predicted that cognitive effort and the message's goal frame
interact with each other to the extent that in the high cognitive effort condition loss
frames would be more effective than gain frames (see hypotheses la and lb). To test
this assumption, a 2 x 2 between-subjects ANCOVA (see Table 4-5) was run on the
data. In contrast to the first study, the analysis revealed no significant main effect for
cognitive effort (F(l, 150) = .00, p > .98). Moreover, the main effect for the message
frame was also insignificant (F(l, 150) = 1.76, p > .18). Consistent with the first
hypothesis, cognitive effort significantly interacted with the message's goal frame
(F(l, 150) = 4.16,p < .05).
As expected, participants in the high cognitive effort condition reported a higher
probability to switch to the upsell option when the offer was loss-framed (Mioss = 4.37)
compared to when it was gain-framed (Mgain = 3.38, t(66) = -2.13, p < .05).
Consequently, participants were more likely to choose the superior car option when
the upsell offer was negatively framed. When, however, the initial cognitive effort was
low, there was no difference between both framing strategies with respect to choice
probability (Mioss = 3.93, Mgain = 4.04, t(85) = .26, p = .80). These results provide
support for hypotheses la and lb and largely replicate the findings from study 1.
Perceived Expensiveness. In order to test hypotheses 2a and 2b, a second ANCOV A
was conducted. The results revealed no significant main effect of cognitive effort on
perceived expensiveness (F(l, 150) = .08, p > .78) and no significant main effect of
70
the message's goal frame on perceived expensiveness (F(l, 150) = .30, p > .58).
Moreover, the interaction between cognitive effort and message frame was marginally
insignificant (F(l, 150) = 2.16, p > .14). Consequently, planned contrasts were not
conducted. Hence, the results could not support hypotheses 2a and 2b. However, the
means for perceived expensiveness illustrated that participants in the high effort
condition rated the price premium for the upsell offer as more expensive when the
offer was gain-framed compared to when it was loss-framed. The results of both
analyses and the corresponding mean values are depicted in Table 4-5 and Figure 4-8.
Table 4-5: Results of the ANCOVAs in Experiment 2
Dependent Variable F(l, 150) p
Gender Choice Probability 3.97 p<.05
(Covariate) Perceived Expensiveness 1.77 p>.18
Cognitive Effort Choice Probability .00 p>.98
Perceived Expensiveness .08 p>.78
Message's Goal Frame Choice Probability 1.76 p>.18
Perceived Expensiveness .30 p>.58
Cognitive Effort x Choice Probability 4.16
p<.05
Message's Goal Frame Perceived Expensiveness 2.16 p>.14
Table 4-6: Mean Values for the Dependent Variables in Experiment 2
ffigh Cognitive Effort Low Cognitive Effort
Loss Frame Gain Frame Loss Frame Gain Frame
Choice Probability 4.37 (1.89) 3.38 (1.90) 3.93 (1.92) 4.04 (1.85)
Perceived Expensiveness 3.90 (1.50) 4.42 (1.67) 4.11 (1.77) 3.93 (1.67)
Anticipated Inaction Regret 4.17 (1.98) 3.26 (1.80) 4.07 (1.82) 3.63 (1.89)
Decision Justifiability 3.90 (2.17) 3.08 (2.10) 3.93 (2.19) 3.35 (2.06)
MC Cognitive Effort Time 72.70 (45.21) 63.92 (53.65) 49.32 (20.28) 50.35 (15.83)
MC Gain Frame 4.97 (1.47) 5.50 (1.03) 5.26 (1.24) 5.74 (1.23)
MC Loss Frame 4.98 (1.32) 4.12 (1.78) 5.20 (1.31) 3.73 (1.82)
Note. The numbers in parentheses represent standard deviations. The manipulation check for cognitive effort
was assessed using decision time measured in seconds. The remaining variables were assessed on seven-point
scales, with higher numbers representing higher mean ratings. MC is the abbreviation for manipulation check.
Figure 4-8: Cognitive Effort and Goal Framing Interactions (Experiment 2)
Choice Probability
4. 5 . .,,,,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
'!<..,,,,, ....... ~ 4.3 7
'!<..,,,,,,,,,,-.::
:-.. .. ~ .... ~,,,,,-.::
~ ... ~ .... ~ .... ~ ........
4. 04 ~ ... ~ ... ~'~"
~ ... ~ ... ~'~"
4 ·;..;;:.,. . .. ~"=············································································
~~'~"
3.93
3.5 .,\. .......................................................................... .
31 T 3:38 '
Low Cognitive Effort High Cognitive Effort
Perceived Expensiveness
5 "'"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
I .,\. .................................................................................................................... .
I 4.11 . 4.42
i ~'''''''"~
4 -1---------------;~-~;--~---~=~'~''''~~,~~~~:~;--~-~~~---------------"
I , .
4.5
3.5
Low Cognitive Effort High Cognitive Effort
..,,,,.Gain Frame
~'''''''''Loss Frame
..,,,,.Gain Frame
~""'"'Loss Frame
71
72
4.3.4.3 Hypotheses Testing: Mediation
In order to investigate the underlying processes m greater depth, two mediation
analyses were conducted.18 Following the recommendations of Baron and Kenny
( 1986), causal steps analysis was employed to examine if anticipated inaction regret
(H3) as well as decision justifiability (H4) mediated the impact of the message's goal
frame on choice probability in the high effort condition. According to this approach, a
variable can be considered as a mediator when (a) the independent variable is a
significant predictor of the intervening variable, (b) the mediator significantly relates
to the dependent variable and ( c) the relationship between the independent and the
dependent variable becomes insignificant (complete) or weakened (partial) once the
mediator is included in the equation (Baron & Kenny, 1986, p. 1176; Preacher &
Hayes, 2004, p. 71 7). Moreover, to formally assess the indirect effect of both
intervening variables, the non-parametric bootstrap analyses based on Preacher and
Hayes (2004; 2008, using the INDIRECT macro) and recommended by Zhao et al.
(2010) was applied. 19 Note that for all analyses, the message's goal frame was dummy
coded. That is, the loss frame was coded "1" and the gain frame was coded "O".
Anticipated Inaction Regret. The first mediation analysis examined if anticipated
inaction regret mediated the impact of the message's goal frame on choice probability
in the high cognitive effort condition. Firstly, the message's goal frame had a
significant impact on choice probability (/J = .25, p < .05). Secondly, the message's
goal frame was also a predictor of anticipated inaction regret (/J = .24, p = .053).
Thirdly, anticipated inaction regret was significantly related to choice probability (/J =
.62, p < .001). Lastly, when both messages' goal frame and anticipated inaction regret
were included in the equation, the mediator remained a significant predictor (/J = .60, p
< .001) while the impact of the independent variable was eliminated (/J = .11, p > .25,
Sobel: z = 1.89, p = .059). These results are also depicted in Figure 4-9. In addition,
employing bootstrap estimation with n = 5000 bootstrap resamples reveals with 95%
confidence that anticipated inaction regret mediated the influence of the message's
goal frame on choice probability with a bootstrap confidence interval [CI] of .0178 to
1.2001 excluding zero. Consequently, hypothesis 3 and the postulated mediation of
anticipated inaction regret are supported.
Decision Justifiability. A second mediation analysis examined separately if decision
justifiability could possibly account for the cognitive process between the message's
18Because the interactive effect of cognitive effort and goal framing on perceived expensiveness was insignificant, no
mediation analysis was performed for perceived expensiveness as the dependent variable.
19For a discussion on both procedures see Preacher and Hayes (2004) and Zhao et al. (2010).
73
goal frame and choice probability in the high cognitive effort condition. Again, a
number of regression analyses were conducted.
The first regression revealed that the message's goal frame was a significant predictor
of choice probability (fJ = .25, p < .05). Moreover, a second regression found a
marginally significant relationship between the message's goal frame and decision
justifiability (fJ = .19, p = .119). Thirdly, decision justifiability was also significantly
regressed on choice probability (fJ = .75, p < .001). Finally, when the message's goal
frame and decision justifiability were both included as predictors of choice probability,
decision justifiability remained significant (fJ = .73, p < .001) while the influence of
the message's goal frame was no longer significant (fJ = .12, p > .16). Sobel' s test (z =
1.56, p = .12) supports the mediating role of decision justifiability in the high cognitive
effort condition with marginal significance. In contrast, bootstrapping narrowly failed
to support these findings (95% confidence interval [CI]: -.1317 to 1.2604).
Consequently, both procedures render mixed support for the mediating role of decision
justifiability and thus for hypothesis 4.
Figure 4-9: Mediation Models in Experiment 2
"'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''\:
~ ~
~ ~
j Anticipated ~
,,------------""'-~! Inaction Regret I
p = . 24 * ,,,,,,-'' ~""""""""""""""""""""""""""""""""'"~ ~= .6 2 * * *
///""' ,,
"'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''"~'''''''''''''''''''''''''''''''" ................................................ ,~,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,')\
~ ~ ~ ~
~ ~ ~ ~ ! Message's Goal ! ! Probability of !
~ ~ ~ Ch . ~ ~ Frame ~"""""""""""""""""""""""""""""""""""""""""""""""~~ oos1ng i
! (O=gain/l=loss) P=.25* , (theUpsell) I
~'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-] ~'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-~
(/J=.11)
~'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'>::
~ ~
~ ~
---------~1 Decision Justifiability I~
,, .. ,,...... ~ ~
P- 19 ,, ' ~ p- 75*** - . ,,,,'' ~""""""""""""""""""""""""""""""""'"' - .
,,,,,,,,''' ~
,,, ............ ,,...... ,$..
"'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-~'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'\; "'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-~"-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-');
~ ~ ~ ~
I Message's Goal I I Probability of I
~ F ~ ~ Ch . ~ ~ rame ~"""""""""""""""""""""""""""""""""""""""""""""""~~ oosmg ~
I (O=gain/l=loss) ! P=.25* ! (theUpsell) !
~ ~ " ~
~ ~ ~ ~
~ ... ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, .. ~ ::.,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,~
(/J= .12)
Note. Relationships apply to the high cognitive effort condition. Values in parentheses indicate the effect of the message's goal
frame on choice probability when the mediator is included; Significance levels * p :S .05; ** p :S .OJ; *** p :S .001.
"Anticipated Inaction Regret" ranges from 1 =regret inaction not at all to 7=regret inaction very much.
"DecisionJustifjphilifJ!." rang.esftom l=morejustifj,able to remain with reservation to 7= more justifiable to switch.
74
4.3.5 Discussion
The purpose of the second study was to replicate the results of study 1 and to test for
the underlying mediation process. A number of important variations were incorporated
in order to enhance the generalizability of the results. That is, experiment 2 was
conducted as an online-study with a balanced consumer sample. Moreover, the study
was adapted to another decision context, namely a car rental choice. In contrast to the
first study, manipulations of the service encounter relied on audio-visual stimuli rather
than written scenarios.
Despite these fundamental changes in the experimental design, the results concerning
hypotheses la and lb could be replicated. It was found that the effectiveness of
different goal frames depended upon the amount of cognitive effort expenditures at the
preliminary decision step. More precisely, it was found that participants who had to
invest a high amount of cognitive effort in the initial reservation task were more likely
to switch to the upsell offer when the sales representative focused on the disadvantages
of the upsell offer. In contrast, participants were more likely to stick with their initial
reservation if they had been confronted with gain-framed arguments. According to the
subsequent mediation analyses this was due to two reasons: First, in the high cognitive
effort condition loss frames prompted the participants to imagine how they would feel
if they missed the upsell offer and thus remained with their initial reservation decision.
Compared to a gain frame, loss-framed offers induced a stronger anticipation of
inaction regret resulting in a greater likelihood of choosing the upsell offer. Second
and closely connected to the aforementioned reasoning, the analysis showed (with
mixed results) that participants found it more difficult to justify inaction (i.e.,
remaining with their reservation) when they were confronted with the disadvantages of
the reservation. Consequently, in order to make a justifiable decision, participants
opted for the upsell option. While loss-framed upsell offers provided strong reasons to
justify a decision shift, gain-framed upsell offers failed to do so when initial cognitive
effort investments were high. As opposed to these results, it was found that different
goal frames did not differ in their effectiveness when initial cognitive effort
investments were low. As such, low cognitive effort built up weaker switching barriers
and thus required less strong upsell arguments. Therefore, gain-framed upsell offers
were also strong enough to persuade the customer of the advantageousness of the
superior car offer.
The results for hypotheses 2a and 2b could not be replicated. That is, the interactive
effect of cognitive effort and message frame on perceived expensiveness was
marginally insignificant. Means for all four conditions did, however, point in the
75
hypothesized direction. This finding must be interpreted in the light of the
heterogonous sample structure. While the sample in the first study consisted of college
students who were largely akin in terms of age and socio-economic background, the
sample in the current study stronger varied in terms of age and income. It can be
assumed that these differences might have accounted for some variance in the
responses. 20
Nonetheless, the study makes three important contributions to this dissertation. First,
the results strengthen the basic assumption that the initial decision process affects
subsequent decision steps. More precisely, the amount of cognitive effort invested at
the initial decision step determines the justifiability of the final decision regarding
acceptance or rejection of the upsell offer. Second, the current study provides
additional support for the differential impact of diverse goal frames. The results reveal
that loss-framed upsell offers provide stronger reasons for a decision switch than gain-
framed upsell offers, especially when initial cognitive effort investments were high.
Third and most important, the study provides a thorough explanation and analysis of
the underlying mediation process. As such, study 2 shows that cognitive effort and
goal framing trigger justification processes. Although high cognitive effort builds up
switching barriers, loss frames provide arguments to overcome these lock-in
tendencies while prompting the decision maker to anticipate feelings of regret. As a
result, anticipation of inaction regret and decision justifiability mediate the relationship
between the message's goal frame and customer's probability of choosing the upsell
option when initial effort investments were high.
Based on these results it seems worth investigating how variations on decision
responsibility alter these relationships. If the feeling of regret is bound to decision
responsibility (see Section 3.2), a lack of responsibility would prevent the customer
from regret anticipation and thus diminish the differential impact of loss- and gain-
framed upsell messages in the high cognitive effort condition. Therefore, one can
propose that decision responsibility or need for justification, respectively, acts as a
boundary condition to the relationships found in study 1 and study 2. Experiment 3
was designed to test this account while directly manipulating need for justification.
20In order to prevent participants from dropping the study, the author refrained from assessing income variables. Therefore,
income could not be included as a covariate. Future studies could, however, directly manipulate the available budget for all
participants in order to control for this potential covariate.
76
4.4 Experiment 3: Decision Justification and Goal Framing
The objective of the third study was to provide direct empirical evidence for the
justification-based explanation of this research. That is, the study was designed to
demonstrate that the differential impact of diverse goal frames in the high cognitive
effort condition was driven by a decision maker's felt need to justify an upsell choice.
Thus, the current study tests the assumption that consumers will be likely to choose the
upsell offer irrespective of the goal frame when they can attribute decision
responsibility to another person (see hypotheses 5a, 5b, 6a and 6b).
4.4.1 Design, Participants, and Procedure
Experiment 3 had a 2 (need for justification: no-justification vs. control) x 2
(message's goal frame: loss vs. gain) factorial design. Both factors were manipulated
between participants and replicated across two different industries (car rental and
hotel) to enhance generalizability. Hypotheses 5 (a, b) and 6 (a, b) are expected to hold
for travel services independently from the type of industry. Therefore, no specific
hypothesis for the influence of the industry has been proposed. The subsequent
analysis revealed no significant interaction between the type of industry and the two
independent variables (ps > .23). Consequently, collapsing responses on the hotel and
car rental scenario seems appropriate for the remaining analyses (for a similar
procedure see Chang, 2008; Inman & Zeelenberg, 2002). Unlike the first two studies,
cognitive effort was kept constant at a high level across all conditions. This was done
because study 1 and study 2 revealed that only in the high effort condition both goal
frames differ in their relative persuasiveness. Accordingly, decision justifiability is
only supposed to account for the differential impact of goal frames when initial effort
investments were high. Thus, incorporating a low effort condition would have
increased the complexity of the experimental design without additional explanatory
contribution.21 A total of 301 undergraduate students participated in the study (average
age of 21.2 years). One third of all participants were female. The paper-and-pencil
experiment was conducted during a lecture at the University of St. Gallen.
Participation in the study was voluntary. All participants were randomly assigned to
one of the four conditions and were entered in a lottery to win one of ten cinema
vouchers to increase motivation. Every participant completed a booklet that contained
the manipulation of the independent variables and the corresponding questionnaire.
21The previous studies showed that, in the low cognitive effort condition, loss and gain frames were equally likely to
persuade the participants of the upsell option. Hence, participants would not need a regret-mitigating justification to opt for
the upsell offer. Manipulating need for justification may have little influence on participants' final choice.
77
The procedure of the study closely followed the proceeding of the first experiment. In
the first part of the study, participants were prompted to imagine that they wanted to
plan their next vacation and needed to reserve either a hotel room (Hotel Charlott) or a
rental car (RentThisCar). Information on the available rooms or cars was provided in
text format to induce a cognitively effortful reservation decision. After the participants
had decided which room or car to reserve they were confronted with a written role-
play scenario that depicted the service encounter. Each scenario manipulated the need
for justification and the message's goal frame, resulting in four different combinations.
For a detailed description of both manipulations see Chapter 4.4.2. Again, at the end of
the scenario the participants had to decide if they would like to opt for the superior car
(room) or if they would rather stick with their initial reservation decision. Moreover,
participants indicated their price perception and answered to several confound and
manipulation checks as well as various covariates. The majority of the participants
needed 15 minutes or fewer to complete the experiment.
4.4.2 Materials and Manipulations
In order to enhance participants' situational involvement and to provide a coherent
understanding, all manipulations were embedded in an introductory story that was
adapted to the realm of students' experiences. Both introductions prompted the
participants to imagine they were going on holiday and they needed a hotel room or a
rental car respectively. Again, fictitious company names were created to heighten
experimental control. In the hotel experiment, participants read:
Soon it will happen - the semester break is approaching! Together with your friends
you plan to travel to Berlin for a long weekend to enjoy the city flair, go shopping and
to party. You have already booked your flight from Zurich to Berlin some time ago.
For the first two nights and days you will be on your own since your friends will still
be working on their last exams. Therefore, you need a hotel room for these nights.
Consequently, you visit the homepage of a booking portal and enter the stay of two
nights as well as your price limit of 50 Euros per night. Finally, you choose a three-
star hotel ("Hotel Charlott'') in the heart of Berlin, from where you can reach most of
the attractions within walking distance. Hence, all you have to do is to decide on one
of the rooms. The following rooms are available in the selected hotel (2 nights/50
Euros per night).
78
In the car rental experiment, the wording of the scenario introduction read as follows:
Soon it will happen - the semester break is approaching! In this year's summer break
for one week, you want to discover Germany and meet some friends and fellow
students on the way. After beginning your road trip in Kiel, you want to rent a car to
drive all the way to Munich (ca. 900 km). You have already booked a flight to Kiel
some time ago. The only thing you have to do is to reserve a rental car, with which you
can drive from Kiel to Munich.
Consequently, you visit the homepage of the car rental company "RentThisCar" and
enter the rental period of one week as well as your price limit of 50 Euros per day.
Hence, all you have to do is to decide which car to reserve. The following screenshot
depicts all options that are available to you.
Both introductions were followed by a screenshot that either illustrated the different
room options or the available rental cars. The information on the different choice
options was arranged in sentence format to trigger enhanced cognitive effort
expenditures and closely followed the high effort manipulation depicted in study 1 and
study 2. The exact cognitive effort manipulations are illustrated in Appendix 3 and 4.
4.4.2.1 Manipulating Need for Justification
Need for justification was manipulated by varying the degree of responsibility for the
final decision on whether to accept or to reject the upsell offer. This procedure was
adapted from Okada (2005) and Mishra and Mishra (2011 ). In particular, participants
were either prompted to imagine that they were joining the service encounter
themselves and had to decide which hotel room or car to take (control condition) or
that a friend was making the choice for them and was therefore responsible for the
final choice (no-justification condition). The rationale behind this manipulation is as
follows: While in the control condition the onus of decision justification is on the
participant, in the no-justification condition choice responsibility and therefore
decision justification will be attributed to the friend (Mishra & Mishra, 2011 ).
Accordingly, when study participants are asked to indicate how the friend should
decide on their behalf, they are likely to indicate a preference that is free from the need
to justify future feelings of regret (Okada, 2005).
The manipulation of the control condition (i.e., need for justification is present)
equaled the decision set-up from the first two studies. That is, participants were
prompted to imagine they arrived at the car rental or hotel themselves to pick up their
car or room key, and that they had to decide upon the upsell offer.
79
In the no-justification condition, however, participants were told that their arrival at
the car rental (hotel) would be delayed and that they had asked a local friend to pick up
the car (room key) and to advance the booking costs. Participants were also informed
that they would refund the rental or booking costs to their friend afterwards in order to
avoid confounding effects that might arise from diverse payment responsibilities.
Consequently, the scenario described that a friend entered the service encounter on
behalf of the study participant and received the upsell opportunity from the counter
employee either in a loss- or in a gain-framed way. Study participants were made
aware that the friend had to make the final choice on their behalf. The exact wording
of the need for justification manipulation in the car rental study is depicted in Figure
4-10 and Figure 4-11. Manipulations for the hotel study can be found in Appendix 4.
4.4.2.2 Manipulating the Message's Goal Frame
Subsequent to the manipulation of need for justification, participants were exposed to
the arguments of the car rental or hotel representative. Due to the paper-and-pencil
character of this study, this was done via written scenarios. Again, the representative
either used loss-framed or gain-framed arguments to convince the participant or the
friend of the enhanced car or room option. Within the car rental and hotel study the
core information remained the same throughout all framing conditions.
That is, in the car rental study the counter employee's arguments referred to (a) the
cabin space and (b) cruise control. While the gain-framed arguments highlighted the
advantages of the superior car offer, loss-framed arguments emphasized the
disadvantages of the reserved car. The wording of both frames was slightly adapted to
fit the need for justification manipulation. In other words, when the friend enters the
service encounter, the salesperson uses fewer personal pronouns than when the
participant received the upsell offer. Both types of offers are illustrated in Figure 4-10
and Figure 4-11.
In the hotel study part the receptionist argues in terms of (a) the view from the room
and (b) the size of the bed. Gain-framed offers were developed so as to highlight the
positive aspects that were associated with the superior room option. In contrast, loss-
framed offers illustrated the negative aspects that were involved with the initially
reserved room. Again, the wording was slightly adapted to the justification
manipulation. The exact manipulation of both frames can be found in Appendix 4.

'/""""""""""""i.
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ = ~
~ .s ~
~ ..... ~
~ ·- ~ ~ ~ ~
~ = ~
~ = ~
~ u ~
~ ~
~ ,.... ~
~ = ~
~ b ~
~ = ~
~ = ~
~ u ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
t ....................................................... J
7,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r,r/i,
~ ~
l Loss Frame j
~ ~
I;,,/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/')!,
The time has come. The semester break has just begun and you
have arrived in Kiel. After you disembarked and collected your
luggage you proceeded to the car rental company where you had
reserved your car. When you finally arrive at the car rental
office you give your name and booking number to the
employee. The car rental representative calls up your reservation
details, checks them and says to you:
"I see that a car from a higher category became available this
morning that I can offer you instead of the car you had reserved.
It would only cost you 10 Euros more per day. The car you had
reserved has two clear disadvantages compared to my offer. The
car you had reserved is smaller and offers you less comfort. I
see that you travel with a lot of luggage. The trunk of the
reserved car is too small to stow your luggage completely. So
you'll have to take some pieces with you in the passenger cabin.
Therefore, there might not be ample space to take friends,
acquaintances or bigger souvenirs with you. Moreover, the car
you had reserved is not equipped with a cruise control.
Accordingly, you can't fzx the tempo. Considering the length of
your journey, this implies that your foot must remain on the
accelerator. Thus, your feet are likely to get weary more
quickly; you will need more stops and will arrive later at your
destination. The car I can offer you has no such disadvantages.
So spare yourself an exhausting car ride and grab this offer.
You will not regret it. "
~/////////////////////////////////////////////////////////////////////:////////////////////////////////////////////////////////////////////////////////////1
! GamFrame !
~ ~
-,.,/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/',/'/;
The time has come. The semester break has just begun and you
have arrived in Kiel. After you disembarked and collected your
luggage you proceeded to the car rental company where you had
reserved your car. When you finally arrive at the car rental
office you give your name and booking number to the
employee. The car rental representative calls up your reservation
details, checks them and says to you:
"I see that a car from a higher category became available this
morning that I can offer you instead of the car you had reserved.
It would only cost you 10 Euros more per day. This offer has
two clear advantages compared to the car you had reserved.
First, it is bigger and offers you more comfort. I see that you
travel with a lot of luggage. The trunk of this car is big enough
to stow your luggage completely. The passenger cabin remains
free for friends and acquaintances or bigger souvenirs.
Moreover, the car I can offer you comprises an integrated
cruise control. With this device you can fzx the tempo.
Especially on this long journey, this allows you to take your foot
off the accelerator. Your feet remain relaxed; you will need
fewer stops and will arrive earlier at your destination. So enjoy
a relaxed car ride and grab this offer. You will be satisfied. "
~
r;Q•
~ ,..
.... .... ..
~
~ -s·
;:::
S" .....
§"
.55) nor a significant interaction effect between both
independent variables (p > .06). Consequently, both manipulations can be considered
successful.
Moreover, the third manipulation check on cognitive effort revealed that all
participants required high amounts of cognitive effort investments in order to come to
an initial reservation decision. The values in each condition were consistently around
the mid-point of the scale. The effort ratings did not differ across conditions (p > . 72).
Table 4-9 provides an overview of the means of all manipulation checks that were
used in the current study.
Coefound Check. Several confound checks examined if different types of goal framing
and the need for justification manipulation affected the comprehensibility of the
different scenarios. In fact, three confound checks revealed that all four conditions
were equally credible (p > .20), comprehensible (p > .92) and realistic (p > .12). Thus,
it can be concluded that the participant's response behavior was not confounded by
these variables.
Covariates. None of the possible covariates emerged as a significant control variable.
Therefore, knowledge, involvement, persuasion knowledge, gender and regret
proneness were excluded from the subsequent analyses.
86
4.4.4.2 Hypotheses Testing: Moderation
Choice Share. In order to test the hypotheses 5a and 5b - that is the assumption that
under no-justification needs the differential impact of loss- and gain-framed messages
on choice will diminish - a first analysis on the main dependent variable was
conducted. Due to the dichotomous nature of the dependent variable a logit analysis
was applied. More precisely, the data were analyzed using a 2 x 2 logistic regression
with need for justification, goal framing and their interaction as categorical
independent variables and choice as the dichotomous dependent measure. The analysis
revealed a significant main effect of the message's goal frame on choice (Wald x2(1,
N= 301) = 15.68,p < .001). This main effect was qualified by a significant interaction
between need for justification and goal framing (Waldx2(1, N= 301) = 5.15,p < .05).
The main effect of need for justification was not significant (Wald x2(1, N = 301) =
1.42, p > .23).22 To follow up on these effects a decomposition of the interaction
within the different levels of need for justification showed the following results.
Figure 4-12: Choice Share of the Upsell Option (Experiment 3)
Choice Share of Upsell Option (in percent)
l!§§GainFrame
~Loss Frame
No-Justification Control
In line with the hypothesis 5b and in accordance with the results from the first two
studies, a chi-square test revealed that in the control condition the percentage of
participants choosing the upsell offer significantly differed between the loss and the
gain frame condition. Concretely, it showed that while only 29.9% of all participants
in the gain frame condition chose the upsell option, 70.1 % were willing to opt for the
upsell when the offer was loss-framed (Wald x2(1, N = 153) = 16.36, p < .001). A
22In general, the model chi square indicated that including the predictors in the model improved the overall fit as the log-
likelihood (-2LL) significantly decreased by a -2LL value of )(3) = 17.72 (p < .001).
87
rather different pattern of results appeared when participants had to imagine a friend
was making the final choice on their behalf. That is, in the no-justification condition -
when decision responsibility for the respondents is diminished - no significant
difference between both framing techniques emerged. For the gain-framed upsell offer,
43.5% of all participants indicated that they hoped their friend would choose the
superior room or car option for them. When confronted with a loss frame, however,
56.5% indicated they hoped their friend would decide for the upsell option. The
difference in the percentage of participants that wanted their friend to opt for the upsell
option was not significant between the different goal frames (Wald x2(1, N = 148) =
.76, p > .38). As expected, the differential impact between loss and gain frames
disappeared when participants attributed their final choice to a friend. Accordingly,
these results provide full support for hypotheses 5a and 5b (see also Figure 4-12).
Choice Probability. In order to enhance comparability of these results with the
outcomes of the first two studies, a second analysis on the continuous choice measure
was conducted. An ANOVA revealed a significant main effect of the message's goal
frame on choice probability (F(l, 297) = 12.92, p < .001), an insignificant main effect
of the justification condition (F(l, 297) = 1.22, p > .27) and a significant interaction
effect between the message's goal frame and need for justification (F(l, 297) = 4.02, p
< .05). To elaborate on the significant interaction, planned contrasts were performed
within both justification conditions. As predicted, when participants had to decide on
their own - that is, they felt a certain need for decision justification - then they were
more likely to choose the upsell offer when it was loss-framed (Mioss = 4.28) compared
to when it was gain-framed (Mgain = 2.95, t(151) = -4.22, p < .001). When, however,
participants passed decision responsibility to a friend, responses to what they wished
their friend would choose for them did not significantly differ between the different
framing styles (Mioss = 4.06, Mgain = 3.69, t(146) = -1.06, p > .29). These results
provide additional support for the hypotheses 5a and 5b and underpin the findings of
the logistic regression (see also Figure 4-13).
Perceived Expensiveness. Finally, a second ANOV A was conducted in order to test for
the joint impact of need for justification and goal framing on perceived expensiveness
of the upsell price. The analysis revealed no significant main effect of need for
justification (F(l, 296) = .32, p > .57) and a marginally significant main effect of the
message's goal frame (F(l, 296) = 2.49, p > .11) on perceived expensiveness. More
importantly, there was a significant interaction effect between both independent
variables (F(l, 296) = 9.28, p < .01). Again, planned contrasts were performed. As
Figure 4-13 shows, when participants were responsible for the final decision, they
88
rated the upsell price as more expensive when the offer was gain-framed (Mgain = 4.43)
compared to when it was loss-framed (Mioss = 3.58, t(151) = 3.18, p < .01). When,
however, a friend was responsible for the final choice the differential impact of loss
and gain frames on perceived expensiveness disappeared. That is, participants' ratings
of the price expensiveness did not vary as a function of the message's goal frame (Mioss
= 4.04, Mgain = 3.77, t(145) = -1.08, p > .28). Hence, H6a and H6b were fully
supported. Table 4-8 and Table 4-9 provide an overview of the results of both
ANOV As and the mean values of all continuous dependent variables.
Table 4-8: Results of the ANOVAs in Experiment 3
Need for Justification
Message's Goal Frame
Need for Justification x
Message's Goal Frame
Dependent Variable
Choice Probability
Perceived Expensiveness
Choice Probability
Perceived Expensiveness
Choice Probability
Perceived Expensiveness
F(l, 297) p
1.22 p>.27
.32 p>.57
12.92 p <.001
2.49 p>.11
4.02 p<.05
9.28 p<.01
Note. One respondent did not answer to the dependent measure of perceived expensiveness. Accordingly, the
degrees of freedom decrease to 296.
Table 4-9: Mean Values for the Dependent Variables in Experiment 3
No-Justification Control
Loss Frame Gain Frame Loss Frame Gain Frame
Choice Probability 4.06 (2.07) 3.69 (2.28) 4.28 (2.03) 2.95 (1.87)
Perceived Expensiveness 4.04 (1.47) 3.77 (1.57) 3.58 (1.57) 4.43 (1.74)
Anticipated Inaction Regret 4.27 (1.66) 4.09 (1.47) 4.68 (1.45) 4.08 (1.49)
Decision Justifiability 3.53 (1.80) 3.19 (1.87) 4.05 (1.88) 3.03 (1.79)
MC Message Frame 5.26 (1.64) 1.88 (1.14) 5.05 (1.49) 2.28 (1.32)
MC Cognitive Effort 3.46 (1.23) 3.55 (1.34) 3.62 (1.37) 3.38 (1.37)
Note. The numbers in parentheses represent standard deviations. All variables were assessed on seven-point
scales, with higher numbers representing higher mean ratings. Due to the dichotomous nature of the
dependent variable choice share and the manipulation check for need for justification, these values were not
included in the table. MC is the abbreviation for manipulation check
89
Figure 4-13: Need for Justification and Goal Framing Interactions (Experiment 3)
Choice Probability
5 "'"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
4.51
t ~'''~'''''''~'''~' 4.28 ~ =--''-''''~''" i 4. 06 $-.'-''~'''''''~'''''''~''''''''IN
3.: 13·~~'
3 +················································································ ........................... .
..,,,,.Gain Frame
~'''''''''Loss Frame
I 2.95
2. 5 ,..,, ... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2 I T '
No-Justification Control
Perceived Expensiveness
5 "'"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
I 4. 5 ., .... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4 t 4:~~ ~:43 -GfilnFnune
I > ~'''''''"" ~""'"'Loss Frame
I 3.77 ~ ~''''''""
' ~''"" I ~,,~ 3.58 3.:c===
No-Justification Control
90
4.4.4.3 Hypotheses Testing: Mediation
Examining the role of anticipated inaction regret and decision justifiability through
mediation analysis was not the primary purpose of the third study. Nonetheless, it
seems appropriate for two reasons. First, study 2 only provided mixed support for the
mediating role of decision justifiability. Although the current study provides direct
empirical evidence for the underlying decision process, it might be interesting to
underpin and strengthen the mediation results from study 2 with another sample.
Second, because the control condition in the current study completely corresponds to
the high cognitive effort condition in study 2 it was feasible to easily apply causal
steps analysis (Baron & Kenny, 1986) and bootstrapping (Preacher & Hayes, 2004,
2008) in the control condition.
Anticipated Inaction Regret. Regression analyses were conducted to examme if
anticipated inaction regret mediated the impact of the message's goal frame on choice
probability and perceived expensiveness when participants had invested much
cognitive effort at the initial choice stage and were responsible for the final decision.
First, the message's goal frame (O=gain frame, l=loss frame) had a significant and
positive impact on choice probability (/J = .33, p < .001) and a significant negative
impact on perceived expensiveness (/J = -.25, p < .01). Second, goal framing was also
significantly related to the anticipation of inaction regret (/J = .20, p < .05). Third,
anticipated inaction regret was also a significant predictor of choice probability (/J =
.29, p < .001) and perceived expensiveness (/J = -.19, p < .05). Lastly, when the
message's goal frame and anticipated inaction regret were both integrated in one
equation to predict choice probability the impact of the independent variable was
reduced (/J = .28, p < .001) whereas the impact of anticipated inaction regret remained
significant (/J = .24, p < .01, Sobel: z = 2.11, p < .05). Moreover, when the message's
goal frame and anticipated inaction regret were put in one equation to predict
perceived expensiveness, the impact of the goal frame was slightly reduced (/J = -.22,p
< .01) whereas anticipated inaction regret remained only marginally significant (/J = -
.15, p = .07, Sobel: z = -1.74, p = .08). Additionally, bootstrapping method according
to Preacher and Hayes (2004, 2008) was applied and underpinned these findings. A
confidence interval of 95% and a number of n = 5000 bootstrap resamples were
specified. With a bootstrap confidence interval excluding zero (.0375 to .4991) the
results support the hypothesis H3 for choice probability (see Figure 4-14). For
perceived expensiveness the confidence interval barely covers zero (-.3031 to .0031)
and therefore fails to support hypothesis 3 for perceived expensiveness.
Figure 4-14: Mediation of Anticipated Inaction Regret in Experiment 3
:>.'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''-.;:
~ ~ ' ~ ! Anticipated ~~
.. ~~ . ~ ,,-----""' I Inaction Regret I
Choice Probability:
p = 29***
91
,...... ~ ~
P = .20* ,,,,,,,,,-'' L"""""""""""""""""""""""""""""""'j ~erceived Expensiveness:
,,,,,,,,,-'' *
---""' P=-.19
,................ ,::.
"'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''"~''''''''''''''''''''''''''''''''\: ,, ... ,,,,,,,,,,,,,,,~ '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''')\
~ ~ ~ ~
~ ~ ~ ~
~ M 'GI~ ~ ~ i essage s oa i i i
~ Frame ~,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,'"'"'"'"'"'"'"'"'"'"'"'"'"~~ Reaction ~
! (O=gain ; 1 =loss) ! Choice Probability: I
L"""""""""""""""""""""""""""""""'J p = . 3 3 ** * (/J = . 2 8 * * *) L"""""""""""""""""""""""""""""""'J
Perceived Expensiveness:
p = -.25** (/J = -.22**)
Note. Relationships apply to the control condition. Values in parentheses indicate the effect of the message's goal frame on
the dependent variable when the mediator is included; Significance levels * p::; .05; ** p::; .OJ; *** p::; .001.
"Anticipated Inaction Regret" ranges from 1 =regret inaction not at all to 7=regret inaction very much.
Decision Justifiability. A second set of regression analyses examined the mediating
role of decision justifiability in the control condition. The regression model revealed a
significant impact of the message's goal frame on choice probability (/J = .33, p <
.001) and perceived expensiveness (/J = -.25, p < .01) as well as on decision
justifiability (/J = .27,p = .001). Moreover, decision justifiability was also significantly
related to choice probability (/J = .58, p < .001) and perceived expensiveness (/J = -.31,
p < .001). More importantly, when the message's goal frame and decision justifiability
were both included as predictors, decision justifiability continued to have a significant
impact on choice probability (/J = .53,p < .001) and perceived expensiveness (/J = -.26,
p = .001) while the impact of the message's goal frame was greatly reduced (choice
probability: p = .18, p < .01, Sobel: z = 3.19, p < .01; perceived expensiveness: p = -
.18,p < .05, Sobel: z= -2.60,p < .01).
Additionally to these findings, bootstrapping also confirmed that decision justifiability
acted as a mediator between the message's goal frame and choice probability in the
control condition, as the 95% confidence interval of the indirect effect did not include
zero ( .2386 to 1.0128, n = 5000 resamples). Similar results can be obtained for
perceived expensiveness as the dependent variable (-.5027 to -.0827, n = 5000
resamples). These results are also depicted in Figure 4-15 and provide support for
hypothesis 4 for both dependent variables. Moreover, the findings strengthen the
results from study 2 by providing clear evidence for the mediational impact of decision
justifiability.
92
Figure 4-15: Mediation of Decision Justifiability in Experiment 3
''-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-"'
~ ~
l l Choice Probability:
)~~ Decision Justifiability ~ p = .58***
p = .27*** ,,,,,,,,,,,,,,,,,,,,,,,,,,,,, t""""""""""""""""""""""""""""""""J s:erceived Expensiveness:
,,,f p - 31 ***
,,,,,,,,,,,,,,,,,'' ," - -.
:'I''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''"~'''''''''''''''''''''''''''''" ·~'''''''''''''''''''''''''''''"~ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
~ ~ ~ ~
~ M 'GI~ ~ ~ ~ essage s oa ~ ~ ~
! Frame ~"""""""""""""""""""""""""""""""""""""""""""""""~ Reaction ' ! (O=gain; 1 =loss) ! Choice Probability: !
~ ~ *** ** ~ ~ '""""""""""""""""""""""""""""""""'"~ p = . 3 3 (/J = .18 ) '""""""""""""""""""""""""""""""""'"'
Perceived Expensiveness:
** * P= -.25 (fl= -.18)
Note. Relationships apply to the control condition. Values in parentheses indicate the effect of the message's goal frame on
the dependent variable when the mediator is included; Significance levels * p::; .05; ** p::; .OJ; *** p::; .001.
"Decision Justifiability" ranges from 1 =more justifiable to remain with reservation to 7= more justifiable to switch.
4.4.5 Discussion
The third study explored the role of need for justification as a boundary condition on
choice and price perception. Following the logic of hypotheses 5 (a, b) and 6 (a, b), it
was predicted that when justification needs were absent or at least diminished, the
differential impact of diverse goal frames would disappear. Put differently, when need
for justification served as a boundary condition, then one could expect that under no
justification needs the differential effect of loss- and gain-framed upsell messages on
customers' probability of choosing the upsell option and perceived expensiveness
should cease to hold. The results of study 3 support these hypotheses and provide
direct evidence for the proposed underlying role of justification processes in the high
cognitive effort condition. That is, it showed that when customers were responsible for
their final decision in the service encounter (control condition) loss-framed messages
could unfold their persuasive strength. More precisely, when confronted with the
potential drawbacks of the reserved car or hotel option, customers anticipated how
they would feel when the service representative was right and were thus more likely to
justify a decision switch compared to when they were confronted with the benefits of
the upsell offer. This finding also underpins the results from study 1 and study 2. A
different pattern of results was found when choice responsibility shifted away from the
participants to a friend (no-justification condition). When asked to indicate what they
hoped their friends would choose on their behalf, participants provided a theoretical
choice that was not driven by a need to justify action or inaction. On that condition,
93
participants were equally likely to choose the loss-framed and the gain-framed upsell
offer. As predicted, this pattern of results also emerged for perceived expensiveness of
the upsell price. In the no-justification condition, participants rated the upsell price
equally expensive irrespective of the goal frame. The arguments provided by each goal
frame contributed equally to the participant's perception of the upsell price. In the
control condition, however, participants experienced the gain-framed upsell offer as
more expensive (compared to the value that was provided by the offer) as opposed to
the loss-framed upsell offer. According to the underlying mediation process,
participants perceived the extra charge for the upsell offer to be less justifiable when
the upsell offer highlighted the possible gains derived from the upsell instead of the
potential losses that were associated with the initial reservation. Under the assumption
that perceived expensiveness might be conceptually reversed to a participant's
willingness to pay (for a discussion see 3 .1 ), it can be concluded that loss-framed
upsell offers provided stronger arguments to justify the expense of extra money (when
initial effort investments were high) compared to their gain-framed counterparts.
At this point it should be recognized that these results emerged for both industries -
hotel and car rental industry - that were collapsed for the conducted analyses.
Nonetheless, it might be important to highlight that separate analyses for both
industries revealed the same and significant pattern of results for the logistic
regression, the ANOV As as well as the mediation analyses.
Besides these promising outcomes, two limitations have to be recognized. First, it
might be argued that the way need for justification was manipulated might lack some
external validity. This may be due to the fact that the actual decision maker viewed the
decision through the eyes of another person (e.g., a friend). While concerns on
restrictive realism are principally eligible, one should note that external validity played
only a minor role in study 3. Instead, it was of considerable importance to uncover the
intrinsic and unconscious process with a stimulus that strongly manipulated need for
justification. Considering this objective, the need for justification manipulation can be
considered successful. Second, in order to reduce the complexity of the experimental
design, cognitive effort was held constant across all conditions. That is, each
participant was exposed to a high effort reservation scenario to induce cognitive lock-
in. This was done because need for justification was only expected to mediate between
goal framing and choice probability (perceived expensiveness) when initial effort
investments were high. In the low cognitive effort condition, however, the impact of
both goal frames did not differ (see study 1 and 2). Consequently, combining the low
effort scenario with the no-justification condition should not have triggered any
94
changes on the outcome variables. However, to provide full support for this assertion,
future studies could also manipulate need for justification in the low effort condition.
To sum up, experiment 3 clearly illustrated that need for justification acts as a
boundary condition when initial effort investments were high. That is, when choice
responsibility on the final decision step shifts to another party (e.g., a friend) the
customer finally decides independently from the message frame. These results provide
strong evidence for the proposed cognitive effort and framing interaction as well as the
underlying justification process. Based on these findings, it seems worth investigating
how the relationship between cognitive effort and message framing changes, when the
initial reservation decision - instead of the upsell choice - is completely delegated to a
commercially acting surrogate who invests a certain amount of cognitive effort in the
initial reservation task. This question will be addressed in the next experiment.
4.5 Experiment 4: Perceived Effort and Goal Framing
The fourth experiment was designed in order to investigate the influence of a
surrogate shopper's initial effort investment on the final choice of the customer. The
objective of the current study was to test for the interactive effect of perceived effort
and the message's goal frame on the customer's willingness to accept an upsell offer
and hence to provide empirical evidence for hypotheses 7a and 7b. Unlike the first and
the second studies, this experiment accounts for decision situations where the customer
completely delegates the initial reservation decision to a surrogate (i.e., travel agent).
Moreover, this study is also distinct from study 3 in that decision surrogacy is applied
to the initial decision step and that the surrogate's decision is manipulated on two
distinct cognitive effort levels.
4.5.1 Design, Participants, and Procedure
The experiment employed a 2 (perceived effort: high vs. low) x 2 (message's goal
frame: loss vs. gain) between-subjects factorial design. A total of 152 undergraduate
students (36.4% females, 63.6% males, average age of 22.5 years) from the University
of St. Gallen participated in the study. The experiment was conducted as a paper-and-
pencil questionnaire during two different classes at the bachelor's and master's level.
Participation was voluntary and all subjects were randomly assigned to one of the four
experimental conditions. To increase task involvement, every participant was offered
the possibility to win one of ten cinema vouchers. In a self-paced manner, participants
could work through the study booklet that contained the stimulus materials as well as
95
the questionnaire. The procedure of the experiment closely followed the lower
decision path from the conceptual model (see Figure 2-2). That is, participants were
prompted to imagine that they were going on holiday and needed a rental car at their
destination. They were informed that, due to a lack of their own experience, they
assigned the complete decision task on which car to reserve to a travel agent. All
participants were asked to immerse themselves into the situation at the travel agency
that was provided to them in a written form. Within this first scenario, a description of
the travel agent's search and decision behavior served as a clue for the participants to
assess the employee's effort investment. This initial scenario resulted in a reservation
choice that was made by the travel agent. After completing a number of filler
questions, participants worked through the second part of the experiment. Therein,
respondents were exposed to the role-play scenario that depicted the service encounter
at the fictitious car rental "RentThisCar" at the second and final choice stage of the
conceptual model. Similar to the previous studies, the car rental representative offered
the participants a superior car option for a premium of 18 Swiss Francs [13 Euros]. He
either employed a loss or a gain frame. Immediately after reading the scenario
participants indicated their choice probability and answered several confound and
manipulation checks. Again, a number of covariates were included in the study.
Processing through the study took between 12 and 15 minutes, depending upon the
effort condition.
4.5.2 Materials and Manipulations
Following the procedure of the previous experiments, a short preliminary frame was
provided to all participants to build up a coherent understanding of the following
scenario. Again, this introduction was adapted to students' interests. Consequently, all
participants read the following introduction that was directly followed by the
manipulation of perceived effort:
Imagine: The end of term is approaching. To reward yourself for the hard work of the
last months, you plan a one-week trip to Germany to meet some of your friends. You
want to start your trip in Berlin, rent a car, visit some cities and then drive back to
St. Gallen. In order to book the flight and to reserve a rental car, you visit a nearby
travel agency and tell the agent of your travel plans. Prior to this, you have already
selected a flight that only needs to be booked now. But when it comes to choosing and
reserving the rental car you need the assistance and expertise of the travel agent. You
have set yourself a price limit of 80 Swiss Francs [58 Euros} per rental day. The travel
agent notes this ieformation, browses for a proper offer in his system and says:
96
4.5.2.1 Manipulating Perceived Effort
In contrast to the first three studies, the current experiment did not manipulate the
cognitive effort exerted by the participant, i.e., the customer; but perceived effort, i.e.,
the customer's perception of the cognitive effort exerted by the employee. Research on
perceived effort showed that the cognitive effort exerted by surrogates is not directly
assessable by consumers (Bechwati & Xia, 2003). Talcing into account that decision
processes of surrogates are generally latent to others, it becomes apparent that "it is not
actual effort that matters to consumers, but perceived effort" (Morales, 2001, p. 2).
Consequently, the manipulation of perceived effort needed to be based on other, more
obvious criteria than the cognitive effort manipulation in the former studies.
Recent publications have shown that when it comes to evaluate the cognitive effort
exerted by others, consumers usually rely on perceptual cues that are easily observable
and trigger awareness to infer the magnitude of others' effort investments (Bechwati &
Xia, 2003; Mohr & Bitner, 1995). According to a qualitative study of Mohr and Bitner
(1995), employee effort is usually inferred from (a) active helping and persistence, (b)
time spent for the customer, and ( c) energy and enthusiasm invested in problem
solving. 23 Research on the duration heuristic underpins the finding that especially time
spent on a service task plays a critical role for effort perceptions, as duration can be
seen as a "carrier of value" (Yeung & Soman, 2007, p. 316). As such, customers
believe that the duration of a service is positively correlated with the effort invested in
service delivery.
According to these results and in line with other studies in the field of employee effort
(Huang, 2010; Mohr & Bitner, 1995; Specht et al., 2007), the travel agent's search and
decision behavior in the current study was manipulated using all three perceptual cues.
That is, study participants were given a script of the travel agent's statements as well
as a description of his behavior in terms of persistence, time and energy. Hence, in the
condition of high perceived effort, the travel agent continuously informed the
participants about his approach to narrow the choice set of available cars (energy),
reviewed the narrowed choice set a second time (persistence) and spent ample time in
the reservation decision (time). Accordingly, the scenario read:
23Besides these three criteria, Mohr and Bitner (1995) additionally presented cues that were strongly associated with the
outcome and quality of effortful employee decisions. Since the focus of the current study lies on the cognitive aspect of
employee effort, only those cues were selected that were closely connected to the conceptual definition of perceived effort.
97
The travel agent buries himself in the reservation decision for some minutes and
steadily ieforms you about the criteria he uses to select the appropriate car. Finally,
he says to you: "I have narrowed the extensive range of available offers to three cars.
But let me peruse the exact car descriptions before I finally reserve one of them. "
Again, he reads through all the offers, compares them and says: "Well, I think, after
all the things you told me about your trip, this car here [he points on his computer
screen} is most appropriate for you and it is the best deal. Besides, it is in your price
range of 80 Swiss Francs a day. When you agree, just give me your signature and I
will reserve the car for you. "
In the low perceived effort scenario, however, the travel agent did not noticeably
narrow the choice set to a few most fitting options (energy), reviewed the options for
just one time (persistence) and decided very quickly which of the cars might have been
most appropriate for the customer's needs (time). Hence, the participants in the low
perceived effort condition read:
The travel agent shortly considers the options and says: "After all the things you told
me about your trip, I can quickly tell which of the available cars is the best deal for
you. I would reserve this car here [he points on his computer screen} - I think you will
like it. Besides, it is in your price range of 80 Swiss Francs a day. When you agree,
just give me your signature and I will reserve the car for you. "
These manipulations were integrated in a short scripted conversation that can be found
in Appendix 5. Note that in none of the conditions did participants receive any detailed
information on the attributes of the available alternative cars. In fact, the travel agent
performed all of the decision making tasks at the initial choice step, beginning with the
definition of the attribute set up to the reservation of one of the rental cars. Particular
attention was given to the wording of both scenarios to ensure that the travel agent in
both effort manipulations was considered equally competent and trustworthy. This was
done because one could expect that perceived incompetence and inefficiency may
counterbalance the effectiveness of the perceived effort manipulation. Mohr and Bitner
(1995) showed that an employee who was perceived as trying hard because of a lack
of competence induced frustration and dissatisfaction among customers. A series of
pretests revealed that the travel agent in both effort manipulations was indeed
perceived as equally competent and trustworthy. Moreover, items for both attributes
were also included in the main study to capture possible confounding effects (see
Chapter 4.5.3.3 for detailed information).
98
4.5.2.2 Manipulating the Message's Goal Frame
Following the manipulation of perceived effort and some filler questions, participants
were confronted with the second written scenario that depicted the service encounter at
the fictitious car rental company "RentThisCar". The participants were informed that
some time has passed and they wanted to pick up the car the travel agent had reserved
for them in advance. Similar to the previous studies, the message's goal frame of the
upsell offer was manipulated using either loss- or gain-framed arguments. In the gain
frame condition, the statements of the representative indicated the benefits gained with
the offered rental car. In the loss-frame condition, however, factually equivalent
arguments stressed the disadvantages of the reserved car. In particular, the rental car
employees' arguments focused on both (a) the cabin space as well as (b) the car
equipment. The exact wording of both manipulations can be found in Appendix 5.
4.5.3 Selection of Measures
4.5.3.1 Dependent Variable
For the fourth study, choice probability served as the mam and only dependent
variable.24 The dependent measure was assessed using the same two-item scale as in
the first and second study. That is, subsequent to the upsell offer of the car rental
representative, participants were asked to indicate "How probable is it that you will
decide in favor of the offer of the car rental representative instead of your
reservation?" (1 =highly improbable, ?=highly probable) and "How certain is it that
you will choose the offered car instead of the reserved car?" (1 =highly uncertain to
?=highly certain). Both items were averaged to form a single index of choice
probability (r = .80), with higher values indicating an increased likelihood of choosing
the upsell option.
4.5.3.2 Covariates
As in the first study, predispositional variables capturing the knowledge about cars,
knowledge about car rentals, persuasion knowledge, situational involvement and
gender were included in the questionnaire in order to control for their influence on the
dependent variable. More precisely, knowledge about cars (r = .66) and car rentals
24Due to missing conceptual and empirical evidence from literature, no hypotheses regarding the mediation effect and the
impact of both independent variables on price perception were derived. Consequently, measures for perceived
expensiveness, decision justifiability, and anticipated inaction regret were not included in the current study.
99
(r = . 72) were each assessed using two items ("I know a lot about ... " and "I have
acquired a lot of experience with ... "). Moreover, persuasion knowledge was assessed
using the six-item scale that has already been applied in the former studies (a= .81).
Finally, participants indicated their level of situational involvement by completing
three statements on their current concentration, attentiveness and involvement (a =
.91 ). Except for the open gender measure, all scales used seven-point items anchored
by strongly disagree (1) and strongly agree (7).
4.5.3.3 Manipulation and Confound Checks
To check for the success of both manipulations - i.e., perceived effort and goal
framing - two manipulation checks were employed. In accordance with recent
publications (Huang, 2008, 2010) perceived effort of the travel agent was captured
applying the five-item seven-point scale adopted from Mohr and Bitner (1995). While
in the original scale two items were reverse-worded, in the current study all items were
formulated in the same direction so as to increase comprehensibility (for a discussion
see e.g., Woods, 2006). Sample items were "The travel agent exerted a lot of energy
when choosing the rental car" or "The travel agent was very persistent". All items
were averaged to form one composite perceived effort index that showed a high level
of reliability (a= .90). Higher index values were supposed to indicate perceptions of
increased employee effort. A second manipulation check was applied to test for the
success of the framing manipulation. These items were adapted from the first study.
Both the gain and the loss frame were examined using two two-item sets. On the first
item set, participants indicated the extent to which the offer of the car rental
representative was based on the negative consequences and disadvantages of the
reserved car (loss frame, r = .66). The second item set, however, rated the extent to
which these arguments focused on the positive consequences and advantages of the
superior car offer (gain frame, r = .73). All items used seven-point scales.
As already addressed under 4.5.2.1, it was of particular importance to assure that the
travel agent in the first scenario was considered equally trustworthy and competent
across both perceived effort conditions. Hence, two items ( "I trust in the travel agent,
that he chose the best offer for me" and "I assume that the decision of the travel agent
was a good one") were included in the study subsequent to the perceived effort
manipulation. Moreover, two additional confound checks served to ensure that the
scenarios were rated as equally credible and comprehensible using the seven-point
semantic differentials from the first study. All measures and reliability coefficients are
summarized in Table 4-10.
100
Table 4-10: Measures employed in Experiment 4
Measures
Dependent Variables
Choice Probability
Covariates
Knowledge Cars
Knowledge Car Rentals
Persuasion Knowledge
Situational Involvement
Gender
.. Af a11ipult1tion <;heclcs ...
Perceived Effort
Message's Loss Frame
Message's Gain Frame
... <;onfo11nd.. <;hecks ...
Trustworthiness
Competence
Comprehensibility
Credibility
4.5.4 Results
Items
2
2
2
6
3
1
5
2
2
Reliability Source
r = .80 Chandran & Morwitz (2005)
r = .66 Mishra et al. (1993)
r =. 72 Mishra et al. (1993)
a= .81 Bearden et al. (2001)
a= .91 Miniard et al. (1991)
a=.90
r= .66
r= .73
Mohr & Bitner (1995)
Meyers-Levy & Maheswaran (2004)
Meyers-Levy & Maheswaran (2004)
Giirhan-Canli & Maheswaran (2000)
Giirhan-Canli & Maheswaran (2000)
4.5.4.1 Checks on Experimental Design
Manipulation Checks. The analyses revealed that both manipulations had been
successful. Participants indicated that the travel agent was highly persistent, energetic
and enduring when they had been confronted with the high perceived effort scenario
(MHighPerceivedEffort = 3.92) as compared to participants in the low perceived effort
condition (MLowPerceivedEffort = 2.40, F(l, 148) = 69.01, p < .001). Additionally, an
ANOV A tested whether the second treatment or the two-way interaction between both
independent variables affected the manipulation check of perceived effort. As
expected, these effects became insignificant (ps > .27).
Moreover, the examination of both framing indices revealed that, first, a loss-framed
upsell message conveyed more negative information (MLossFrame_LossCond = 5.80) than a
gain-framed upsell offer (MLossFrame_GainCond = 4.10, F(l, 148) = 71.39, p < .001).
Second, participants also felt that a gain-framed upsell message conveyed more
positive aspects (MGainFrame_GainCond = 5.91) than when the message was loss-framed
(MGainFrame_LossCond = 4.68, F(l, 148) = 41.17, p < .001). Again, no other treatment
effects were significant (ps > .29). Table 4-12 depicts the means for both manipulation
checks.
101
Coefound Checks. To check for potential undesired effects of the manipulations on
other variables, a number of ANOV As were performed. The results suggest that the
evaluation of the travel agent between both perceived effort conditions did not differ in
terms of trustworthiness (p > .54) and perceived competence (p > .97). Moreover,
participants rated the depicted scenarios as equally credible (p > .55) and
comprehensible (p > . 72) throughout all four conditions.
Covariates. Finally, the hypothesis test revealed that gender was a significant covariate
(p < .01). No other potential covariates reached statistical significance. Accordingly,
these variables were excluded from the subsequent analysis.
4.5.4.2 Hypothesis Testing
A 2 x 2 ANCOV A tested the assumption that when surrogates invested lower amounts
of effort at the initial choice stage, customer's probability of choosing the upsell offer
varies as a function of the message's goal frame. In other words, hypotheses 7a and 7b
postulated that perceived effort and the message's goal frame interact. The results
support these assumptions. That is, the analysis showed a marginally significant main
effect of perceived effort (F(l, 146) = 2.60, p = .109) and a significant main effect of
the message's goal frame (F(l, 146) = 16.93, p < .001) on choice probability. More
importantly, the ANCOV A revealed a significant interaction effect between both
independent variables (F(l, 146) = 6.65,p < .05).
According to the logic of the seventh hypothesis, planned contrasts were performed
within both perceived effort conditions. When participants noticed that the travel agent
had invested much cognitive effort in the reservation decision, framing the upsell
message in terms of losses or gains did not yield any significant differences on choice
probability (Mgain = 2.89, Mioss = 3.35, t(72) = -1.25,p > .21).
However, a different pattern of results emerged for those conditions, where perceived
effort investment was low. In these cases, participant's likelihood of choosing the
upsell offer was significantly determined by the goal frame of the upsell offer. Thus,
participants were more likely to choose the upsell offer when the car rental
representative referred to the disadvantages of the reserved car (Mioss = 4.26) instead of
the advantages of the superior new offer (Mgain = 2.81, t(76) = -4.14, p < .001). Table
4-11 and Figure 4-16 summarize the results of the ANCOV A and the mean values of
the planned contrasts.
102
Table 4-11: Results of the ANCOVA in Experiment 4
Dependent Variable F(l, 146) p
Gender (Covariate) Choice Probability 9.72 p<.01
Perceived Effort Choice Probability 2.60 p=.109
Message's Goal Frame Choice Probability 16.93 p <.001
Perceived Effort x Choice Probability 6.65
p<.05
Message's Goal Frame
Table 4-12: Mean Values for the Dependent Variables in Experiment 4
High Perceived Effort Low Perceived Effort
Loss Frame Gain Frame Loss Frame Gain Frame
Choice Probability 3.35 (1.66) 2.89 (1.49) 4.26 (1.69) 2.81 (1.38)
MC Loss Frame 5.95 (.85) 3.99 (1.54) 5.66 (1.00) 4.21 (1.33)
MC Gain Frame 4.79 (1.46) 6.01 (.76) 4.59 (1.55) 5.81 (.85)
MC Perceived Effort 3.79 (1.19) 4.03 (1.24) 2.49 (.96) 2.33 (1.01)
Note. The numbers in parentheses represent standard deviations. All variables were assessed on seven-point
scales, with higher numbers representing higher mean ratings. MC is the abbreviation for manipulation check.
Figure 4-16: Perceived Effort and Goal Framing Interaction (Experiment 4)
Choice Probability
......,.Gain Frame
3 ~"""'~Loss Frame
2.81 ... ..... 2.89
2.5
Low Perceived Effort High Perceived Effort
103
4.5.5 Discussion
The fourth and last study of this dissertation examined hypotheses 7a and 7b. The
experiment investigated the interactive effect between perceived effort and the
message's goal frame on the probability that the customer chooses the upsell offer.
Contrary to the first three studies, the experiment focused on the lower conceptual path
of the prefixed model - that is, on situations where the initial reservation decision was
completely delegated to a surrogate shopper. This experimental variation provided
insights to the effect of a surrogate shopper's decision on the subsequent decision of
the customer. The results support the general assertion that customers take heed of the
cognitive effort exerted by others on their behalf.
Compared to study 1 and study 2, the fourth study showed a reversed pattern of results
regarding the interplay between perceived effort and goal framing.
The results show that when the surrogate shopper completely incurred the reservation
decision and noticeably invested much cognitive effort in this choice process, the
customer becomes locked-in to the decision of the surrogate. Considering that the
customer did not know the initial choice set, this lock-in was supposed to be more
powerful than the lock-in situation in the previous experiments. In line with this
reasoning, the results showed that framing the subsequent upsell offer either in terms
of losses or in terms of gains did not yield any differential customer reactions.
When, however, the customer realized that the surrogate invested only a low amount
of effort on his behalf, loss-framed upsell arguments provided an effective tool to
release the customer from the lock-in state. More precisely, participants were more
likely to opt for the upsell offer when the upsell message was framed in terms of losses
as opposed to messages that were framed in terms of gains.
It seems plausible that this pattern of results can be attributed to the underlying
justification process. When the customer decides whether to choose the upsell offer or
not he or she will weigh the available reasons against each other. In the high perceived
effort condition, it seems likely that the arguments in favor of the initial reservation
decision - that is the effort investment of the surrogate and the complete decision
delegation - outweigh the anticipated inaction regret that is induced by a loss-framed
upsell offer. In the low perceived effort condition, however, loss frames are likely to
trigger justification processes that are strong enough to overcome the lock-in, which is
primarily due to the customer's dependence upon the surrogate's reservation decision.
While the results of the experiment are quite encouraging and enhance the
understanding of the impact of surrogate's decisions, the fourth experiment was not
104
designed so as to provide empirical evidence for the underlying mediation process of
decision justifiability and anticipated inaction regret. Instead, the findings can be
interpreted as a supplement to the first three experiments and a first and exploratory
attempt to explore the role of perceived employee effort in customers' decision
processes.
In order to fully understand the effect of others' effort investments on customer's
decision processes in service encounters, future studies could incorporate two
variations. First, in order to fully compare the influence of cognitive effort with
perceived effort on the customer's final choice, a future study could manipulate both
variables in one single experiment. Second, to determine the role of the proposed
mediators, one could replicate the design of the second study for the perceived effort
scenario. Further considerations on future research directions will be part of the
following general discussion.
105
5 General Discussion
As the concluding chapter of this dissertation, the following sections will present a
summarizing discussion of the results and a presentation of its implications. The
remainder of this chapter is structured as follows: With regard to the prefixed research
questions, the first section of this chapter will briefly review the results of the four
experimental studies. On a more generalized level, the second and third section will
elaborate on the theoretical and managerial implications of this dissertation. While the
fourth part will briefly discuss the limitations of the conducted studies, the last and
final section will reveal avenues for future research.
5.1 Summary of Results
The overall objective of the present dissertation was to examme how consumers
respond to upsell offers and why they react in the way they do. While there is missing
empirical evidence on customers' behavioral consequences of upsell offers, the current
research addressed this void by prepending a set of four research questions. The
subsequent paragraph will answer these questions while summarizing the results of the
four experimental studies.
Research Question 1: How and why does the initial product or service choice of
the customer influence the subsequent decision of whether to accept or to refuse
the upsell offer? More specifically, to what extent does the initial cognitive effort
investment influence the upsell choice?
In order to answer this question, the current research conceptualized the customer's
decision process as a three-step choice process comprising the initial reservation
decision, the upsell offer and the final decision regarding acceptance or refusal of the
upsell offer. The conceptual framework proposed that the initial cognitive effort
investments determine the customer's affinity, i.e. the lock-in, to the reserved product
or service option. The first study provides clear evidence for this effect and answers
the question of how initial decisions influence subsequent decisions. It was found that
high initial cognitive effort investments lead to a lock-in of the customer that impedes
switching to the superior and more expensive upsell offer. Compared to low cognitive
effort investments, customers were less likely to switch to an upsell offer when
cognitive effort expenditures at the initial choice step were high. Likewise, it also
showed that these effort investments influenced participants' willingness to pay for the
106
upsell offer and thus the perceived expensiveness of the upsell price. After exerting
high amounts of cognitive effort at the initial choice step, study participants were less
likely to additionally exert financial effort. Accordingly, they were less willing to pay
the price premium and perceived the given upsell price as more expensive than
participants with lower prior effort investments. Study 2 and study 3 reveal why this is
the case. In light of extensive cognitive effort investments it seems to be less
justifiable to switch to another product or service option. In order to prevent cognitive
sunk costs appearing, participants were likely to give priority to their initial choice and
to refuse the upsell.
Research Question 2: Are different selling arguments likely to moderate the
impact of the initial product or service choice on the subsequent decision? Put
differently, which message frame attenuates or intensifies the influence of the
initial cognitive effort investment on the upsell choice?
The findings of this dissertation suggest that the goal frame that is used to
communicate the upsell offer does moderate the impact of the initial choice on the
final decision of the customer. More precisely, the first two experiments, which varied
in terms of sample structure, manipulation technique and selected industry, provide
unequivocal evidence that loss and gain frames are varyingly effective in persuading
the customer of a decision switch. As such, participants in the high cognitive effort
condition, who were more likely to become cognitively locked-in to their reservation
decision, opted for the upsell offer when the associated message was loss-framed (vs.
gain-framed). In other words, loss-framed selling arguments helped to reduce lock-in
and attenuated the impact of initial effort investments while gain frames intensified the
influence of the lock-in on the final decision. However, in the low cognitive effort
condition, both goal frames were equally effective and did not yield varying reactions
of the participants.
A similar pattern of results also emerged for willingness to pay as the dependent
variable. The first experiment demonstrated that, in the high cognitive effort condition,
participants were willing to pay a higher premium for the upsell when the offer was
loss-framed (vs. gain-framed). Again, in the low cognitive effort condition, both
framings did not differ in their relative impact on willingness to pay. For the
dependent measure of perceived expensiveness, which was included as a negative
correlate of willingness to pay, these results emerged in the first study, but not in the
second study. The results of the first experiment demonstrate that, after high cognitive
107
effort exertion, participants perceived the given upsell price to be more expensive
when the upsell message was gain-framed (vs. loss-framed). It was reasoned that
participants in the high cognitive effort condition needed stronger arguments to self-
justify a decision switch towards the upsell offer. As such, it was assumed that loss-
framed upsell offers provided these compelling arguments as opposed to their gain-
framed counterparts. This argumentation leads to the answer of the third research
question:
Research Question 3: Which process variables, i.e., mediators, explain the
customers' choice or refusal of the upsell offer?
As assumed in the conceptual model, studies 2 and 3 found that the variables that
mediated the relationship between the upsell message's goal frame and the final
decision of the customer are anticipated inaction regret and decision justifiability.
These results apply to the high cognitive effort condition. Drawing on decision
justification theory it was argued that participants are likely to decide in favor of the
option that is supported by the best overall reason in order to mitigate anticipated
regret. As such, they were expected to choose the upsell offer, when the message
triggered anticipated inaction regret that might pertain to loss-framed arguments.
Experiment 2 showed that, in the high cognitive effort condition, loss frames were
indeed more likely to trigger the anticipation of regret that might result from sticking
with the reservation (inaction) which, in turn, led to a higher probability of choosing
the upsell. As such, when confronted with a loss frame (vs. gain frame), participants
found it more justifiable to switch to the upsell offer instead of remaining with the
initially reserved option.
This pattern of results was also underpinned by the findings of study 3, which was
designed to provide direct empirical evidence for the proposed justification-based
explanation. Assuming that, first, decision justifiability drives the examined
relationships and, second, decision justification is linked with decision responsibility,
one could expect that the differential impact of diverse goal frames vanishes when
need for justification shifts away from the customer. As such, it was hypothesized that
need for justification in terms of decision responsibility acts as a boundary condition
for the interaction between cognitive effort and goal framing. The results of the third
experiment provide strong evidence for this reasoning. When need for justification was
absent, meaning that responsibility for the final choice was shifted from the participant
to a friend, participants were equally likely to choose the upsell option, no matter
108
which kind of framing was applied. More precisely, when the onus of self-justification
was not on the participants, loss- and gain-framed upsell offers were equally likely to
persuade the participants because the need to mitigate anticipated inaction regret was
absent. This pattern of results also emerged for perceived expensiveness and could be
replicated across two different service industries (hotel and car rental). These results
provide direct evidence for the mediating role of decision justifiability.
Furthermore, additional mediation analyses in the control condition (need for
justification is present) strengthened and extended the results from the second
experiment. As such, study 3 additionally showed that decision justifiability also
mediated the relationship between goal framing and perceived expensiveness when
initial effort investments were high.
Research Question 4: How does the decision process change when the customer
delegates the initial reservation decision to a surrogate shopper? More precisely,
is the cognitive effort exerted by a surrogate likely to act as a moderator?
The fourth study of this dissertation provides initial insights into the influence of
surrogate shoppers. The results of the last experiment clearly demonstrated that the
cognitive effort exerted by a surrogate, such as a travel agent, does act as a moderator
in the subsequent decision process of a customer. In particular, substituting the initial
decision of the customer by a surrogate who decided on behalf of the customer
changed the prior pattern of results. It was assumed that choice delegation per se
creates a lock-in and that the customer is likely to continue with the decision of the
surrogate due to a lack of choice set knowledge. It was proposed, however, that this
lock-in might be reinforced by the level of perceived effort. If the surrogate had
invested much cognitive effort in the reservation decision, it seemed likely that the
customer remained locked-in, irrespective of the goal frame of the upsell offer. In the
reverse case, when the surrogate had invested lower amounts of cognitive effort, loss-
framed (vs. gain-framed) upsell messages were likely to provide strong arguments to
switch to the upsell offer. These predictions were supported. That is, in the high
perceived effort condition, neither the gain nor the loss frame yielded differential
customer reactions. Participants were likely to remain with the initial reservation
decision, irrespective of the goal frame. In the low perceived effort condition,
however, loss frames were likely to release the customer from the lock-in as opposed
to gain-framed messages. Although it was not directly tested, it may be assumed that
these results can also be attributed to the underlying justification process.
109
5.2 Theoretical Contributions
The results of all four studies have diverse implications for theory building and
research in specific areas of consumer behavior. While these implications have been
explicitly discussed in the previous study sections, the current paragraph outlines the
theoretical contributions of this dissertation on a more superordinate and general level.
Hence, the following sections will discuss the impact on the literature on upselling,
and the implications for research on cognitive effort, goal framing and decision
justification.
5.2.1 Contribution to Literature on Upselling
Upselling, as it was conceptualized in this dissertation, is an important tool of a firm's
revenue management. Upselling enables a service provider to manage capacities
profitably and to control inventory when demand is uncertain (Kimes, 2000). Past
research on revenue management predominantly focused on optimization and
forecasting aspects as well as the economic perspective of diverse revenue
management instruments (see e.g., Biyalogorsky et al., 2005; Wirtz et al., 2003).
Although it becomes apparent that, in the end, the customer determines the success of
these instruments, only few publications deal with the behavioral consequences of
diverse practices such as overbooking and price fencing (von Wangenheim & Bayon,
2007; Wirtz et al., 2003). To date, no publication focuses on the determinants of
successful upsell options from the perspective of consumer behavior.
This dissertation addresses this gap by developing a choice model that illustrates the
decision steps in a typical upselling process from the customer's perspective.
Therefore, it explicates the customer's decision process and places the decision maker
in the focus of attention. Moreover, this model outlines how sequential decision steps
interact with each other. It shows that antecedent reservation decisions affect final
upsell choices. Additionally, this dissertation investigates the underlying cognitive
process as well as the factors that trigger decision justifiability. To sum up, the
experimental studies disclose the variables that will support traditional forecasting
models to depict the decision maker's behavior more precisely. Thereby, this
dissertation adds to the growing body of literature that places the customer in the
center of revenue management activities.
Moreover, this dissertation also adds to the upselling literature that deals with the long-
term perspective of customer relationship management. While most of the research in
the field of customer relationship management (e.g., Kim & Kim, 1999; Ngobo 2005)
110
solely focused on relationship-dependent variables between the customer and the
service provider (e.g., frequency of interaction) as well as individual characteristics of
the decision maker (e.g., age, education), the current research highlights the role of
personal interaction within the service encounter. More specifically, the conceptual
model of this dissertation shows that variables aside from long-term customer-seller
relationships affect the upselling potential. Accordingly, the experimental studies shed
new light on upselling as an elaborated sales tool.
5.2.2 Contribution to Research on Cognitive Effort
This dissertation also strives to make a contribution to the literature that predominantly
focuses on cognitive effort as the key determinant variable in consumer decision
behavior (Bettman et al., 1991; Garbarino & Edell, 1997).
First, this dissertation contributes to the understanding of cognitive effort as a driver of
switching costs (Burnham et al., 2003; Cunha & Caldieraro, 2009, 2010; Fornell,
1992). That is, the experimental studies of this dissertation show that own cognitive
effort investments create a lock-in situation that forces the decision maker to stick to a
decision. So far, lock-in situations that stem from cognitive effort investments have
primarily been demonstrated for repeated consumption situations (Johnson et al., 2003;
Murray & Haubl, 2007). According to these authors, lock-in occurs because cognitive
costs associated with choosing and using a product or service decrease over time.
While this lock-in is due to efficiency gains, the current research provides evidence
that cognitive lock-in also occurs for first time decision processes because the
customer commits to an effortful choice in order to spare future effort investments and
to prevent from sunk cognitive costs (similarly see Cunha & Caldieraro, 2009;
Fitzsimons, 2000). It is one of the first systematic investigations of cognitive effort as
an investment. Therefore, the present studies provide initial evidence for a sunk cost
effect that is based on cognitive expenditures, instead of monetary and temporal
investments.
Second, the first three studies of this dissertation underpin cognitive fit theory (Vessey,
1991 ), which provides a theoretical basis for the effectiveness of different information
presentation formats (Speier, 2006). The present studies showed that, as compared to
text format, matrix representations facilitate decision tasks in a given choice set of
about four different alternatives. More specifically, the checks on the cognitive effort
manipulation revealed that different kinds of information presentation formats required
the decision maker to invest different amounts of effort in order to compare the
111
available service options and to come to a reservation decision. While a matrix
presentation induced low amounts of effort investments, organizing the same
information by means of listed texts induced significantly higher effort expenditures.
These results provide promising avenues for future manipulations of cognitive effort.
Finally, this dissertation adds to the understanding of how cognitive effort
expenditures of others impact a customer's choice process. Literature in the domain of
perceived effort has mainly examined the antecedents and emotional consequences of
perceived employees' effort expenditures (Bechwati & Xia, 2003; Huang, 2008; Mohr
& Bitner, 1995). However, this dissertation broadens the knowledge about the
cognitive consequences of perceived effort by showing that customers can become
locked-in to decisions of others. It sheds light on a practically highly relevant and
theoretically neglected sequential choice process where the initial decision step is
completely delegated to a surrogate buyer. Therefore, the present dissertation provides
a first empirical attempt to demonstrate that people can become locked-in to decisions
of surrogate shoppers.
5.2.3 Contribution to Research on Goal Framing
This dissertation also makes an important contribution to the literature on framing and
goal framing in particular. Goal framing is, compared to other framing techniques,
quite new to the framing literature (Levin et al., 1998). Most of the studies in this
research field deal with health-related issues in terms of disease detection and disease
prevention (Banks et al., 1995; Maheswaran & Meyers-Levy, 1990; Moorman & van
den Putte, 2008). However, a growing body of literature applies loss and gain frames
to consumer choice and advertising topics (Ganzach & Karsahi, 1995; Putrevu, 2010;
Roggeveen et al., 2006). These publications investigate the persuasive strength of
different goal frames in non-risky choice settings. This dissertation contributes to this
research stream by applying goal framing to a non-risky, low involvement choice
situation. Moreover, it provides a first highly relevant sample application of loss and
gain frames in a concrete service setting, both theoretically and practically. This new
kind of application required the introduction of a process explanation that accounted
for the multi-level character of decision processes in services. That is, while most of
the research on the effectiveness of loss and gain frames builds upon the negativity
bias (Levin et al., 1998), the present research adopts a so far neglected and more
affect-driven explanation of the effectiveness of diverse goal frames by introducing
anticipated inaction regret as mediating variable. More precisely, the present studies
reveal that loss frames trigger anticipated feelings of inaction regret which in turn
112
affects justification processes. It was shown, however, that contrary to the ratio of the
negativity bias, loss-framed messages are not superior per se. Instead, the goal frame
of an upsell offer interacts with prior cognitive effort investments. That is, loss frames
unfold their persuasive strength when prior decisions require strong arguments. High
cognitive effort investments at the initial choice stage of an upselling process develop
switching barriers that can only be overcome by persuasive arguments that highlight
the negative feelings associated with inaction. This process explanation was confirmed
in two studies and for two different experimental designs and can be considered
important for future studies in the field of goal framing. In this respect, the present
dissertation makes a key contribution to theory development in goal framing research.
Lastly, from a methodological point of view the present studies overcome the so far
ambiguous manipulation problem that is often related to goal framing studies (for a
discussion see e.g., O'Keefe & Jensen, 2006). While some of the above cited studies
used a pure cross-complement design for the framing procedure (see Section 4.2.2.2)
others combine different kernel states with each other, thereby impeding the
comparability of results. The studies of this dissertation, however, consistently apply a
pure cross complement goal framing in order to enable future meta-analytic reviews to
clearly classify the study design and compare the results to other studies.
5.2.4 Contribution to Research on Decision Justification
This dissertation clearly contributes to research on the decision justification theory that
was proposed by Connolly and Zeelenberg in 2002. This theory highlights the role of
decision justifiability and regret in decision making and was used as an explanatory
approach to account for earlier findings on regret on a post-hoc basis (Connolly &
Zeelenberg, 2002). Those reinterpretations and additional studies were partly
counterfactual with past research and theorizing on the "status quo" bias that was
predicted by norm theory (for this theory see Kahneman & Miller, 1986; Reh &
Connolly, 2010; Samuelson & Zeckhauser, 1988). According to this bias, consumers
tend to opt for the status quo because switching induces higher levels of regret than
inaction (Inman & Zeelenberg, 2002; Kahneman & Miller, 1986). However, recent
studies on decision justification (Connolly & Reh, 2003; Reh & Connolly, 2010;
Zeelenberg et al., 2002) weaken the robustness of this bias and suggest that "whether
action or inaction leads to more regret depends on a person's assessment of the
justifiability of the option chosen rather than on whether it involved action or inaction
per se" (Reh & Connolly, 2010, p. 1416). The present studies are in line with these
findings. It shows that customers tend to choose the upsell option when there are good
113
and strong reasons that support a decision switch. Otherwise they remain with their
initial reservation decision. Put differently, inaction (i.e., remaining with the initial
reservation decision) is only the default option when switching is not justifiable.
Obviously, these results are conflicting with predictions of norm theory and provide
full support for the mediating role of anticipated inaction regret and thus for decision
justification theory.
Moreover, this dissertation also contributes to this literature stream by showing that
decision justifiability varies between loss and gain frames depending on the amount of
prior effort investments. Whereas previous studies have only accounted for the impact
of affective prior experiences (positive vs. negative) on decision justification (Inman &
Zeelenberg, 2002; Zeelenberg et al., 2002), the present research has found that
cognitive prior experiences also trigger justification processes. Therefore, the present
results extend the current findings on context effects that might account for differences
on decision justifiability.
5.3 Managerial Contributions
The results of this dissertation have a number of implications for managers especially
in the service industry. These implications reach from an operative to a strategic level
and will be discussed in the following sections.
5.3.1 Implications for Communication Strategies
The first implication of the present findings is straightforward: The success of an
upsell offer depends upon the customer's cognitive effort investment in the reservation
decision and the wording of the upsell offer in the service encounter. Put differently,
the effectiveness of loss- or gain-framed selling strategies varies as a function of the
amount of cognitive effort the customer has spent on prior decision steps. For
managers it seems important to realize that both variables - cognitive effort and
framing - can be controlled to a great extent. Therefore, managers are well advised to
actively design their communication strategy in order to ensure the success of their
upselling attempts. This means that, at a strategic level, firms should synchronize their
communication at the different touch points. That is, the personal communication at
the service counter has to fall into line with the requirements of a preceding
reservation contact (e.g., via internet or phone). For example, an upsell communication
strategy based on a loss frame should be applied when other communication channels,
such as online reservation systems, require an elaborated effort investment of the user.
114
At a more operative level, managers are well advised to design reservation systems
that are usable and highly structured in order to simplify the customer's reservation
decision and to prevent the customer from becoming locked-in. However,
implementing these requirements in a firm's daily business is often challenging for
two reasons. First, an operations manager who is responsible for the counter business
in a hotel or car rental usually might not have the authority to alter the online
appearance of the company. Second, service availability is often subject to major
capacity fluctuations. An increased availability of different service categories (e.g.,
hotel rooms) increases the choice set and thus the required effort investments of the
customer on a daily basis. Both reasons require the employees at the service counter to
flexibly adapt their communication strategy. The present research findings provide a
clear instruction on how to customize the wording of the upsell offer in the service
encounter.
Moreover, practitioners could also use the insights from the present studies to segment
the customers in two different ways and to adjust the communication strategy
according to the target group. As such, service firms might differentiate between
customers who reserved a flight or room either on their own (e.g., booking via online
reservation system) or who delegated this decision to a surrogate such as a travel agent
(e.g., booking via a professional booking device). In the case of a surrogate decision
the customer is likely to be locked-in to the reserved service option. This lock-in
additionally increases with the level of perceived effort. In contrast, customers who
reserved on their own - all else being equal - will only be locked-in to a high-effort
reservation decision. While loss frames are likely to release the customer from this
lock-in in the latter scenario, both framing strategies are equally (in)effective when
travel agents invested high amounts of cognitive effort. Finally, customers could also
be segmented by their level of expertise. Recent studies on the cognitive differences
between experts and novices found strong evidence for a negative relationship
between domain-specific knowledge and cognitive effort needed to process new
information (Ho, 2011; Kalyuga et al., 2003; Sweller, 1988; Wiley, 1998). Although it
was not directly tested in the present studies one can assume that individual expertise
in the specific domain of, for example, booking a flight or a rental car influences
cognitive effort needed to come to a reservation decision (see also Chapter 5.5.1).
Accordingly, different communication strategies at the service counter might prove
successful. Because service companies often hold information on the booking history
of a customer, segmenting the customers according to their expertise might be easily
applicable.
115
5.3.2 Implications for Pricing
The results of the present studies also suggest implications for the pricing of the upsell
offer at the service counter. In general, the results of the first study show that
customers are likely to pay a higher extra charge for an upsell option when they
invested a lower amount of cognitive effort at preliminary choice stages. Put
differently, they rate the price premium of the superior product or service as more
expensive when they invested an increased level of cognitive effort in the reservation
decision. Because this effect is moderated by the message's goal frame, marketers can
amplify customers' willingness to pay when they employ loss-framed upsell messages
after high cognitive effort expenditures. Again, as a basis for price discrimination,
marketers can employ customer characteristics such as "self-booker vs. travel agent"
or "novices vs. experts" to tap the individual upselling potential of each customer.
5.3.3 Understanding Sequentiality and Justifiability of Upsell Decisions
At a more general level, the results of the theoretical and empirical analyses of the
present dissertation underline the process character of customers' decisions in service
settings. Hence, this dissertation contributes to raise a service provider's awareness of
the sequential nature of upselling decisions and a customer's intrinsic need to justify
an upsell choice.
Although in managerial practice there is widespread agreement that the bulk of
purchase decisions is made at the counter or point of sale (GfK, 2009), marketers
should be aware that, nonetheless, these decisions might be the result of choices and
experiences made prior to the service encounter. This dissertation strongly focusses on
the impact of initial decision steps on final decisions and provides empirical evidence
for the importance of prior decisions on subsequent decisions in a service setting.
Moreover, this research may help managers to understand that along this decision
process each decision step brings forth different reasons in favor and against switching
to the upsell option. Therefore, managers should be aware that every customer feels a
certain need to justify a product or service choice towards him- or herself and that this
justification process guides the final decision. Consequently, the customer will only
choose the upsell offer when the arguments in favor are strong enough to overcome the
reasons against the superior offer. This underlying principle has not only caused the
present study results; it seems plausible that it might also be transferable to other
contextual factors that have an effect on the customer. As such, one can assume that
other variables at the initial choice stage, as for example the payment method (for a
116
discussion see Chapter 5.5.1), might bring forth reasons against a decision switch.
Accordingly, other arguments at the service counter, which for example involve the
comparability of offerings (for a discussion see Chapter 5.5.2), are likely to provide
strong reasons in favor of a decision switch. All these variables may unfold their
persuasive appeal through their impact on decision justifiability. Managers who are
aware of the fact that customers' decisions are guided through justification processes
become enabled to adapt their selling strategy flexibly.
5.4 Limitations
As every experimental investigation, this research bears some limitations primarily in
terms of external validity. While those limitations that especially account for the
respective experiment have been discussed in the corresponding sections, the
following paragraphs elaborate on general restrictions that call for future research.
First, it can be noted that three of the four studies relied on student samples. As already
pointed out in 4.1, the appropriateness of this sampling approach has given rise to a
substantial body of theoretical discourse (cf. Hooghe et al., 2010; Peterson, 2001;
Winer, 1999). While some of these researchers argue that results from student samples
might lack external validity and thus generalizability to non-student populations, other
authors acknowledged the appropriateness of student samples for a specified set of
research goals. That is, research that attempts to apply theory across a variety of real-
world situations solely requires the study sample to be homogeneous instead of strictly
representative (Calder et al., 1981). This dissertation clearly focused on developing
and testing a theoretical model that explains how prior decisions affect subsequent
decisions to choose a superior product or service through justification processes.
Accordingly, employing student samples that were homogeneous on non-theoretical
variables such as age and education can be referred to as appropriate for the set
research goals. Moreover, the second study used a balanced adult sample and
replicated the results of the first study that employed a student sample. This
additionally provides evidence that in the context of this dissertation a convenience
sample based on students is likely to provide the same results as a representative
sample of customers.
Second, another limitation of the present studies anses from the experimental
procedure. The sequence of both scenarios, that is the initial reservation decision and
the subsequent decision in the service encounter, was designed to depict the relevant
decision steps in a typical choice process within a service setting. As a consequence, a
117
time span of several days or months (in the real world) was compressed into an
experiment of approximately 15 minutes. While this direct sequence of decision steps
does not necessarily mimic the way consumers decide in real environments, time-
compressed methodologies like these have proven reasonably valid in depicting
decision processes (cf. Burke et al., 1992; Soman, 2001a). However, one may argue
that, in reality, the time gap between the initial reservation decision and the final
choice of the customer significantly impacts the upsell choice. This may be due to two
reasons. First, there is evidence that over time people forget about their decisions,
especially difficult ones (Chance & Norton, 2008). This would imply that customers
may be less likely to recall the initial reservation decision and the associated cognitive
effort investment when the temporal gap between both decision steps increases.
Accordingly, the cognitive lock-in effect would diminish over time. Second, research
on mental accounting suggests that consumers track temporal and monetary costs
asymmetrically across different accounting periods (Soman, 2001b; Soster et al.,
2010). As such, temporal costs are likely to be written off more quickly than monetary
costs. To date, there is missing evidence on how individuals track cognitive costs.
However, because time and cognitive effort are interlinked concepts (see Section
4.2.3.3) one can assume that cognitive effort may also be written off at the end of an
accounting period and would therefore not be considered at following decision steps.
Hence, in order to investigate the boundary function of a time gap, future studies could
apply a longitudinal design to capture the moderating influence of the decision period.
Third, it might also be of interest to investigate the influence of different upsell prices.
The present studies used a price premium of approximately 20 percent for an upsell to
the next service category. This rate followed from a comparison of customary market
prices in the hotel and car rental industry. Future studies may serve to show that the
findings of this dissertation are not limited to this price increase. Instead one can
assume that the relationship between cognitive effort and message frame may be
strengthened when the service employee offers a more expensive product or service
from a superordinate service class because need for justification would increase.
Finally, one should notice that a generalization of the study results beyond the context
of this research (i.e., hotel or car rental service setting) may be limited. The results of
the present experiments show that the predicted relationships hold for two different
service industries, namely the hotel and the car rental industry. Both branches share
the following key characteristics: (a) as opposed to financial or insurance services they
involve low potentials of financial or physical risk; (b) they provide the customer the
right to use a product or service for a defined period of time and ( c) require a
118
reservation beforehand. Consequently, services that accomplish these criteria include
car rentals, hotels, passenger transportation and theater businesses. However, for future
studies it might be worth investigating the factors that determine the decision process
in more risky, high-involvement services like banking, legal and insurance services.
5.5 Further Research and Extensions
The model proposed in the present dissertation highlights the influence of cognitive
effort at an initial decision step on a customer's decision to opt for an upsell offer and
the moderating impact of the upsell message's goal frame. The model was, however,
not constructed as a comprehensive account for consumer decisions in upselling
situations; rather it depicted those variables that were assumed to account for a large
part of variance in consumer choice in upselling contexts. Additionally, in order to
address the aforementioned limitations and to outline avenues for future research a
number of extensions can be incorporated in the model. The following sections will
briefly summarize these starting points for further research.
5.5.1 Lock-in Factors
The current dissertation argues that the effort investment at the initial choice step plays
a crucial role for the subsequent decision on whether to accept an upsell offer or not. It
was shown that high cognitive effort expenditures lead to a lock-in situation. However,
one might argue that beside the cognitive effort expenditure, lock-in effects may also
be triggered by other factors. These factors, that can also be termed "lock-in factors",
will be discussed in the following passages.
Advance Payment. Services such as booking a hotel room or a rental car are often
subject to an advance or prepayment. Here, the customer is offered the opportunity to
pay for the service prior to the time of service acquisition in return for a price discount
(Shugan & Xie, 2004). Compared to a cash-and-carry payment option, where time of
acquisition and payment coincide (Soman & Lam, 2002), prepayment would take
place at the initial reservation step of the proposed conceptual model of this
dissertation. Research on mental budgeting has shown that prior financial spending in
one product category influences future spending in the same category (Heath & Soll,
1996; Soman & Lam, 2002). More precisely, Heath and Soll (1996) demonstrate that
past expenses reduce the amount and likelihood of future spending because decision
makers actively keep track of past expenses and set budgets for each expense category.
All else being equal, one can assume that, in the context of the present research, a
119
prepayment option at the initial choice stage acts like a cognitive effort investment. It
seems likely that a customer becomes financially locked-in to a reservation decision
that was associated with a prepayment. Switching to a more expensive upsell option
may seem unjustified unless arguments provide strong reasons to support the choice of
the upsell option. Presumably, loss-framed upsell offers may provide these arguments
because they allow the decision maker to anticipate the regret he or she would sustain
if the initial reservation decision went wrong. Accordingly, switching to the upsell
option and thus mitigating anticipated inaction regret is likely to outweigh the pain of
an additional payment. However, gain-framed upsell offers seem to lack this
persuasive power when the customer had to prepay the service. Similar to the low
cognitive effort condition in the present research - all else being equal - one could
assume that both frames would not differ for the cash-and-carry payment option.
Expertise. As already pointed out under 5.3.1, it seems likely that the level of cognitive
effort investment needed to accomplish a decision task may be strongly associated
with the domain-specific knowledge (i.e., expertise) of a customer. A number of robust
research findings support this notion. It appears that domain-specific knowledge enters
the long-term memory of a person in a hierarchically and categorized schema
organization (Chi et al., 1981; Kalyuga et al., 2003). Existence and elaboration of these
schemas primarily distinguish between experts and novices (Sweller, 1988). While
highly elaborated schemas allow experts to "follow well-traveled [ ... ] solution paths"
in information processing and problem solving (Hong & Stemthal, 2010, p. 301),
novices have to rely more heavily on external information search (Beatty & Smith,
1987). This suggests that customers with elaborated domain knowledge need to exert
lower amounts of cognitive effort than less knowledgeable customers (see Ho, 2011
for a first empirical evidence). Accordingly, future studies could use domain-specific
expertise of the customer as a proxy for cognitive effort. This would allow for the
conclusion that novices are in general more likely to become cognitively locked-in to
their initial reservation decision. Accordingly, loss- and gain-framed upsell offers may
unfold their differential persuasive impact especially for novices as opposed to experts.
As such, extending future research attempts to the influence of prior knowledge at the
initial decision step would yield concrete segmentation implications for the managerial
practice.
120
5.5.2 Contextual Factors of the Service Encounter
It seems obvious that there are a number of additional factors that can be assumed to
moderate the relationship between effortful initial decisions of the customer and the
upsell offer of the service representative. These factors, also referred to as "contextual
factors" of the service encounter, can either relate to the characteristics of the upsell
message (e.g., attribute alignability), the characteristics of the service employee (e.g.,
source credibility) or the characteristics of the service setting (e.g., social context) (for
a similar classification see Stiff & Mongeau, 2003).
Attribute Alignability. While the present studies examined the influence of different
wordings of the upsell offer, that is different goal frames; further research may wish to
investigate the moderating impact of the content of the message itself. More precisely,
it seems worth investigating which product differences should be made salient in order
to persuade the customer from the upsell offer. Based on the research on attribute
alignability, two products of one product class can be compared either through
alignable or nonalignable attribute differences (Markman & Medin, 1995; Zhang &
Fitzsimons, 1999; Zhang & Markman, 1998). Alignable differences arise from
products that differ on a commonly shared product dimension. Otherwise,
nonalignable differences exist, when these products differ on a unique, unshared
attribute (Bertini et al., 2009). The concept of attribute alignability can be easily
adapted to the context of upsell offers. According to the definition, an upsell offer
presents a product or service option with enhanced performance attributes. These
enhancements can be either based on alignable, more similar attributes or on
nonalignable, more dissimilar properties. On the one hand, "alignable enhancements"
refer to improvements on attributes, the reservation and the upsell option share in
common. On the other hand, "nonalignable enhancements" refer to the attributes that
are unique for the upsell offer (for a similar conception see Okada, 2006). For
example, an alignable room offer could refer to the increased bed size as compared to
the bed size in the reserved hotel room. Instead, a nonalignable offer would outline
that, for example, the enhanced room comprises a multimedia system that is not
available in the reserved room. There is significant evidence that nonalignable
differences between products are cognitively more taxing than alignable comparisons
since they require across-attribute trade-offs (Bertini et al., 2009; Johnson, 1984). It
also shows that decision makers are more likely to remember alignable differences
(Zhang & Markman, 1998). More importantly, decision makers seem to base the
justification of their decisions primarily on alignable instead of nonalignable attributes
(Markman & Medin, 1995). This empirical evidence would imply that nonalignable
121
attributes would be less compelling for the customer because they additionally
increase cognitive effort and do not lend themselves as a source of justification to
cancel out already invested cognitive effort. Consequently, one could assume that after
high cognitive effort investments, customers are less likely to choose a nonalignable
upsell offer as compared to an alignable upsell offer.
However, a competing explanation introduced by Okada (2001, 2006) would postulate
a quite distinct pattern of results. Motivated by the question on how to convince
owners of existing products of upgrading to a new and enhanced product, Okada
(2006) tested the impact of alignable and nonalignable product enhancements on
upgrading decisions. Applying a mental accounting approach she hypothesized that
customers will only opt for an enhanced product when this product is dissimilar to the
existing one. In a series of four studies, she showed that nonalignable product
enhancements allow the customer to open another mental account that is disassociated
with the existing mental account and the corresponding sunk costs of the initial
financial investment. Assuming that cognitive effort also enters a mental account as a
negative entry, one could postulate that nonalignable upsell offers would be more
persuasive than alignable ones because the sunk costs that would arise from a
switching decision would be disassociated with the upsell offer. Future studies could
test which of these two competing explanations might apply in the present research
context.
Source Credibility. Another promising avenue for future research would be to include
a variable that does not only account for the message content but also for the
customer's impression of the salesperson who offers the upsell to the customer. A
salesperson characteristic that has received remarkable research attention in the last
years is source credibility (for a review see Pompitakpan, 2004). Source credibility
refers to whether a customer perceives a salesperson, advertiser or company as
competent and trustworthy (Pompitakpan, 2004; Sharma, 1990). There is considerable
evidence that, compared to low-credibility sources, highly credible message providers
are considered to be more persuasive and influential (e.g., Pompitakpan, 2004;
Sharma, 1990). However, Tormala and colleagues (2006) provide a first empirical
evidence for a reversal of the credibility-persuasiveness link that may be interesting in
the current context. In both experiments the authors show that the impact of a source's
credibility depends upon the strength of arguments. More precisely, when confronted
with a strong message, participants considered a highly credible source as more
persuasive than a low credible one. When the arguments of the message were weak,
however, high source credibility backfired and induced less favorable attitudes than
122
low source credibility. The authors argue that, according to the underlying self-
validation hypothesis, perceived source credibility determines the message receiver's
confidence in information validity and thus persuasion (Tormala et al., 2006). High
argument quality intensified this effect - low argument quality attenuated the link
between source credibility and persuasiveness. Similar processes may appear in the
present research context when, in addition to the effort and goal frame manipulation,
source credibility would be manipulated. It seems possible that, after high effort
investments, the influence of loss-framed upsell messages (strong arguments) on
choice may be weakened when the information on the low credibility of the
salesperson follows the upsell message. In contrast, relatively weak arguments, such as
gain-framed upsell offers, would gain persuasive power when the salesperson was
assessed less credible. Imagine, for example, that after an effortful initial reservation
decision the salesperson at the car rental counter offers you an upsell option. The
salesperson uses loss-framed arguments that finally give you strong reasons to
positively respond to his offer. Before you tell him your decision you look at the wall
behind the counter and see a picture of this particular salesperson. You realize that he
is the branch manager. The short description below his picture shows you that he has
been working for the car rental for two decades. This information may suggest to you
that the salesperson is highly credible, which might strengthen your decision in favor
of the upsell option. Due to the credibility-reversal-effect one could assume that less
compelling arguments such as gain-framed messages might gain argumentative
strength when source credibility is low. Although these arguments seem plausible,
there is no empirical evidence for a three-way interaction between initial effort
investments, goal framing and source credibility. Therefore, future studies should
account for the influence of a salesperson's credibility in order to allow for a
comprehensive understanding of the attenuating and reinforcing role of source
characteristics in upselling situations.
Social Context. From a theoretical as well as practical perspective it might also be
interesting to investigate how the reported effects may be influenced by the social
context. As such, future studies could investigate how the mere presence of other
customers affects the target customer's upsell decision in the service encounter and
how different kinds of social groups (e.g., peers, family, and colleagues) moderate the
influence of cognitive effort expenditures and the message's goal frame. Past research
shows that, in a retail or service context, decision makers act and decide differently
when other social entities are merely present (Argo et al., 2005; Zhou & Soman,
2003). Research has found that as the size and physical closeness of a noninteractive
123
social group increases (e.g., other customers in a queue) customers are likely to
activate self-presentation concerns (Argo et al., 2005). That is, customers are likely to
acquire products or services in order to gain approval by others, to symbolically
represent a group membership (Leigh & Gabel, 1992) or to present themselves in a
positive light (Wooten & Reed II, 2004). These results may have the following
implication for the present research context: When other customers enter the service
counter, the target customer will be likely to bolster his impression management.
Compared to a situation without social presence, it seems likely that strong self-
presentation concerns backfire. Because a customer may not want to appear regretful
he or she is more likely to refuse a loss-framed upsell offer when other customers are
present. In contrast, gain-framed upsell offers may become more persuasive, because
the customer may be tempted to show a certain financial strength in the presence of
others. These effects are, however, likely to be moderated by individual differences on
the self-representation style as well as the norms of the social group (Wooten & Reed
II, 2004). Future research could even try to push the envelope further and investigate
the impact of different sources of social influences exerted by friends, family or
colleagues. There is initial evidence that these sources differently impact a customer's
final decision and willingness to pay (Childers & Rao, 1992; Kurt et al., 2011).
5.5.3 Macro-level Factors of Upsell Offers
At a superordinate level, two variables can finally be distinguished that may also
determine the success of an upsell offer - namely dynamic and cultural aspects.
Dynamic Aspects. The main focus of the present research was on the behavioral and
short-term consequences of upsell offers. In other words, the current studies
investigated the influence of diverse goal frames on customer's choice and perceived
expensiveness (willingness to pay). However, in order to ensure the long-term success
ofupsell options future research should also consider dynamic and long-term effects of
different selling techniques. More precisely, future studies could account for the
influence of the effort investment and message frame on proximate decisions. It seems
likely that both the decision process at the service counter as well as customer
experiences with the service afterwards influence future decision processes. It might
be interesting to investigate customers' satisfaction with the service encounter and the
chosen service option after using the service or product. As such, future research
might ask the following questions: Would a customer's satisfaction with a loss-framed
upsell option that turned out to be a good choice outweigh the satisfaction of a gain-
framed upsell option? Conversely, would it be more detrimental to future businesses
124
when a loss-framed upsell option, instead of a gain-framed one, turned out a failure?
How would these service experiences influence reservation decisions und willingness
to pay in the future? In order to answer these questions, future studies could employ a
longitudinal experimental design in a real service setting.
Culture. Service encounters, especially in the travel industry, are becoming
increasingly international (World Trade Organization, 2011 ). In so called "inter-
cultural service encounters", where service providers and customers come from
different cultural backgrounds (Stauss & Mang, 1999), being sensitized towards
culture-specific customer expectations and behaviors is imperative for a service firms
upselling success. The growing literature on cross-national service experiences
provides evidence that culture constitutes a framework for the interaction between
service employees and customers (Mattila, 1999; Zhang et al., 2008). It shows,
however, that "social rules and customer expectations that are related to service
encounters are likely to vary from culture to culture" (Mattila, 1999, p. 376).
Recent cross-cultural research, which primarily builds upon Hofstedes' (1991) cultural
dimensions, gives rise to the expectation that the strength of the initial lock-in as well
as the effectiveness of diverse goal frames may vary between cultures. As such,
research on restaurant switching showed that customers from collectivistic cultures
(e.g., Taiwanese) seek to express individuality by switching behaviors more strongly
than U.S. customers who are stronger bound to individualistic cultural values (Lin &
Mattila, 2006). However, an initial study on switching barriers in service settings
yielded no significant differences between cultures (Patterson & Smith, 2003). Based
on these contradictory results, further research may wish to investigate the strength of
the cognitive lock-in and its effect on the final decision between different cultures.
In addition, the following insights might become important for the framing of the
upsell offer in inter-cultural service encounters. There is initial evidence that, for
example, German customers have lower expectations on the service quality compared
to customers from the United States (Witkowski & Wolfinbarger, 2002). Moreover,
while Western cultures favor direct and explicit communications, Eastern cultures are
especially prone to non-verbal communications (Hall, 1984). Thus, it seems likely that
Asian cultures may react more negatively to the explicit offer of an upsell. In contrast,
one could expect that American customers anticipate becoming "upsold", because
upselling is a widely established sales tool in many industries in the United States.
Consequently, testing different framing techniques in a cross-cultural setting would
add explanatory power to the present study results in order to upsell - not to upset -
the customer in inter-cultural service encounters.
125
6 References
Aggarwal, P. (1998). Deciding Not to Decide: Antecedents of Consumer Decision
Delegation. Dissertation, Syracuse University, Syracuse, NY.
Aggarwal, P., & Mazumdar, T. (2008). Decision Delegation: A Conceptualization and
Empirical Investigation. Psychology & Marketing, 25(1), 71-93.
Ahluwalia, R., Burnkrant, R.E., & Unnava, H.R. (2000). Consumer Response to
Negative Publicity: The Moderating Role of Commitment. Journal of
Marketing Research, 37(2), 203-214.
Albarracin, D., & Wyer, R.S., Jr. (2000). The Cognitive Impact of Past Behavior:
Influences on Beliefs, Attitudes, and Future Behavioral Decisions. Journal of
Personality and Social Psychology, 79(1), 5-22.
Argo, J.J., Dahl, D.W., & Manchanda, R.V. (2005). The Influence of a Mere Social
Presence in a Retail Context. Journal of Consumer Research, 32(2), 207-212.
Arkes, H.R., & Blumer, C. (1985). The Psychology of Sunk Cost. Organizational
Behavior and Human Decision Processes, 35(1), 124-140.
Baek, T.H., & Morimoto, M. (2012). Stay Away from Me: Examining the
Determinants of Consumer Avoidance of Personalized Advertising. Journal of
Advertising, 41(1), 59-76.
Banks, S.M., Salovey, P., Greener, S., Rothman, A.J., Moyer, A., Beauvais, J., & Epel,
E. (1995). The Effects of Message Framing on Mammography Utilization.
Health Psychology, 14(2), 178-184.
Baron, R.M., & Kenny, D.A. (1986). The Moderator-Mediator Variable Distinction in
Social Psychological Research: Conceptual, Strategic, and Statistical
Considerations. Journal of Personality and Social Psychology, 51(6), 1173-
1182.
Bateson, J.E.G., & Hui, M.K. (1992). The Ecological Validity of Photographic Slides
and Videotapes in Simulating the Service Setting. Journal of Consumer
Research, 19(2), 271-281.
Bauer, H.H., Hammerschmidt, M., & Braehler, M. (2003). The Customer Lifetime
Value Concept and Its Contribution to Corporate Valuation. Yearbook of
Marketing and Consumer Research, 1, 47-67.
Bearden, W.O., Hardesty, D.M., & Rose, R.L. (2001). Consumer Self-Confidence:
Refinements in Conceptualization and Measurement. Journal of Consumer
Research, 28(1), 121-134.
126
Beatty, S.E., & Smith, S.M. (1987). External Search Effort: An Investigation across
Several Product Categories. Journal of Consumer Research, 14(1), 83-95.
Bechwati, N.N., & Siegal, W.S. (2005). The Impact of the Prechoice Process on
Product Returns. Journal of Marketing Research, 42(3), 358-367.
Bechwati, N.N., & Xia, L. (2003). Do Computers Sweat? The Impact of Perceived
Effort of Online Decision Aids on Consumers' Satisfaction with the Decision
Process. Journal of Consumer Psychology, 13(112), 139-148.
Bergen, M., Dutta, S., & Walker, O.C., Jr. (1992). Agency Relationships in Marketing:
A Review of the Implications and Applications of Agency and Related
Theories. Journal of Marketing, 56(3), 1-24.
Berry, L.L., Wall, E.A., & Carbone, L.P. (2006). Service Clues and Customer
Assessment of the Service Experience: Lessons from Marketing. Academy of
Management Perspectives, 20(2), 43-57.
Bertini, M., Ofek, E., & Ariely, D. (2009). The Impact of Add-on Features on
Consumer Product Evaluations. Journal of Consumer Reserach, 36(1), 17-28.
Bettman, J.R., Johnson, E.J., Luce, M.F., & Payne, J.W. (1993). Correlation, Conflict,
and Choice. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 19( 4 ), 931-951.
Bettman, J.R., Johnson, E.J., & Payne, J.W. (1990). A Componential Analysis of
Cognitive Effort in Choice. Organizational Behavior and Human Decision
Processes, 45(1), 111-139.
Bettman, J.R., Johnson, E.J., & Payne, J.W. (1991). Consumer Decision Making. In T.
S. Robertson & H. H. Kassarjian (Eds.), Handbook of Consumer Behavior (pp.
50-84). Englewood Cliffs: Prentice Hall.
Bettman, J.R., & Park, C.W. (1980). Effects of Prior Knowledge and Experience and
Phase of the Choice Process on Consumer Decision Processes: A Protocol
Analysis. Journal of Consumer Research, 7(3), 234-248.
Bhatnagar, A., Misra, S., & Rao, H.R. (2000). On Risk, Convenience, and Internet
Shopping Behavior. Communications of the ACM, 43(11), 98-105.
Bitner, M.J. (1990). Evaluating Service Encounters: The Effects of Physical
Surroundings and Employee Responses. Journal of Marketing, 54(2), 69-82.
Bitner, M.J., Booms, B.H., & Tetreault, M.S. (1990). The Service Encounter:
Diagnosing Favorable and Unfavorable Incidents. Journal of Marketing, 54(1),
71-84.
127
Biyalogorsky, E., Gerstner, E., Weiss, D., & Xie, J. (2005). The Economics of Service
Upgrades. Journal of Service Research, 7(3), 234-244.
Bloch, P.H., & Richins, M.L. (1983). A Theoretical Model for the Study of Product
Importance Perceptions. Journal of Marketing, 47(3), 69-81.
Blodgett, J.G., Hill, D.J., & Tax, S.S. (1997). The Effects of Distributive, Procedural,
and Interactional Justice on Postcomplaint Behavior. Journal of Retailing,
73(2), 185-210.
Bohutinsky, C.H. (1990). The Sell up Potential of Airline Demand. M.S. Thesis,
Massachusetts Institute of Technology, Cambridge.
Bolton, R.N., Lemon, K.N., & Verhoef, P.C. (2008). Expanding Business-to-Business
Customer Relationships: Modeling the Customer's Upgrade Decision. Journal
of Marketing, 72(1), 46-64.
Burke, R.R., Harlam, B.A., Kahn, B.E., & Lodish, L.M. (1992). Comparing Dynamic
Consumer Choice in Real and Computer-Simulated Environments. Journal of
Consumer Research, 19(1), 71-82.
Burnham, T.A., Frels, J.K., & Mahajan, V. (2003). Consumer Switching Costs: A
Typology, Antecedents, and Consequences. Journal of the Academy of
Marketing Science, 31(2), 109-126.
Burnkrant, R.E., & Sawyer, A.G. (1983). Effects of Involvement and Message Content
on Information Processing Intensity. In R. J. Harris (Ed.), Ieformation
Processing Research in Advertising (pp. 43-64). Hillsdale, NJ: Lawrence
Erlbaum.
Cacioppo, J.T., & Gardner, W.L. (1999). Emotion. Annual Review of Psychology, 50,
191-214.
Calder, B.J., Phillips, L.W., & Tybout, A.M. (1981). Designing Research for
Application. Journal of Consumer Research, 8(2), 197-207.
Campbell, M.C., & Kirmani, A. (2000). Consumers' Use of Persuasion Knowledge:
The Effects of Accessibility and Cognitive Capacity on Perceptions of an
Influence Agent. Journal of Consumer Research, 27(1), 69-83.
Cardozo, R.N. (1965). An Experimental Study of Customer Effort, Expectation, and
Satisfaction. Journal of Marketing Research, 2(3), 244-249.
Celsi, R.L., & Olson, J.C. (1988). The Role of Involvement in Attention and
Comprehension Processes. Journal of Consumer Research, 15(2), 210-224.
128
Chance, Z., & Norton, M.I. (2008). Decision Amnesia: Why Taking Your Time Leads
to Forgetting. Paper presented at the Association for Consumer Research,
Memphis, TN, USA.
Chandran, S., & Morwitz, V.G. (2005). Effects of Participative Pricing on Consumers'
Cognitions and Actions: A Goal Theoretic Perspective. Journal of Consumer
Research, 32(2), 249-259.
Chang, C. (2008). Ad Framing Effects for Consumption Products: An Affect Priming
Process. Psychology & Marketing, 25(1), 24-46.
Cheema, A., & Soman, D. (2006). Malleable Mental Accounting: The Effect of
Flexibility on the Justification of Attractive Spending and Consumption
Decisions. Journal of Consumer Psychology, 16(1 ), 33-44.
Chemev, A. (2003). When More Is Less and Less Is More: The Role of Ideal Point
Availability and Assortment in Consumer Choice. Journal of Consumer
Research, 30(2), 170-183.
Chi, M.T.H., Feltovich, P.J., & Glaser, R. (1981). Categorization and Representation
of Physics Problems by Experts and Novices. Cognitive Science, 5(2), 121-152.
Childers, T.L., & Rao, A.R. (1992). The Influence of Familial and Peer-Based
Reference Groups on Consumer Decisions. Journal of Consumer Research,
19(2), 198-211.
Connolly, T., & Reh, J. (2003). Omission Bias in Vaccination Decisions: Where's the
"Omission"? Where's the "Bias"? Organizational Behavior and Human
Decision Processes, 91(2), 186-202.
Connolly, T., & Zeelenberg, M. (2002). Regret in Decision Making. Current
Directions in Psychological Science, 11(6), 212-216.
Cooper-Martin, E. (1994). Measures of Cognitive Effort. Marketing Letters, 5(1), 43-
56.
Cornelissen, G., Dewitte, S., Warlop, L., & Yzerbyt, V. (2007). Whatever People Say I
Am, That's What I Am: Social Labeling as a Social Marketing Tool.
International Journal of Research in Marketing, 24(4), 278-288.
Cunha, M., Jr., & Caldieraro, F. (2009). Sunk-Cost Effects on Purely Behavioral
Investments. Cognitive Science, 33(1), 105-113.
Cunha, M., Jr., & Caldieraro, F. (2010). On the Observability of Purely Behavioral
Sunk-Cost Effects: Theoretical and Empirical Support for the BISC Model.
Cognitive Science, 34(8), 1384-1387.
129
Donaldson, C., Thomas, R., & Torgerson, D.J. (1997). Validity of Open-Ended and
Payment Scale Approaches to Eliciting Willingness to Pay. Applied Economics,
29(1), 79-84.
Du, J., Fan, X., & Feng, T. (2011). Multiple Emotional Contagions in Service
Encounters. Journal of the Academy of Marketing Science, 39(3), 449-466.
Fennema, M.G., & Kleinmuntz, D.N. (1995). Anticipations of Effort and Accuracy in
Multiattribute Choice. Organizational Behavior and Human Decision
Processes, 63(1), 21-32.
Fiedler, K. (2007). Construal Level Theory as an Integrative Framework for
Behavioral Decision-Making Research and Consumer Psychology. Journal of
Consumer Psychology, 17(2), 101-106.
Field, A. (2009). Discovering Statistics Using SPSS (Vol. 3). Thousand Oaks: Sage.
Fiske, S.T., & Taylor, S.E. (1984). Social Cognition. Reading, MA: Addison-Wesley.
Fitzsimons, G.J. (2000). Consumer Response to Stockouts. Journal of Consumer
Research, 27(2), 249-266.
Folkes, V.S. (1984). Consumer Reactions to Product Failure: An Attributional
Approach. Journal of Consumer Research, 10(4), 398-409.
Fornell, C. (1992). A National Customer Satisfaction Barometer: The Swedish
Experience. Journal of Marketing, 56(1), 6-21.
Friestad, M., & Wright, P. (1994). The Persuasion Knowledge Model: How People
Cope with Persuasion Attempts. Journal of Consumer Research, 21(1), 1-31.
Frijda, N.H., Kuipers, P., & ter Schure, E. (1989). Relations among Emotion,
Appraisal, and Emotional Action Readiness. Journal of Personality and Social
Psychology, 57(2), 212-228.
Gabaix, X., & Laibson, D. (2006). Shrouded Attributes, Consumer Myopia, and
Information Suppression in Competitive Markets. Quarterly Journal of
Economics, 121(2), 505-540.
Gabbott, M., & Hogg, G. (2000). An Empirical Investigation of the Impact of Non-
Verbal Communication on Service Evaluation. European Journal of Marketing,
34(314), 84-98.
Ganzach, Y., & Karsahi, N. (1995). Message Framing and Buying Behavior: A Field
Experiment. Journal of Business Research, 32(1), 11-17.
130
Garbarino, E.C., & Edell, J.A. (1997). Cognitive Effort, Affect, and Choice. Journal of
Consumer Research, 24(2), 147-158.
Garbarino, E.C., & Slonim, R. (2003). Interrelationships and Distinct Effects of
Internal Reference Prices on Perceived Expensiveness and Demand.
Psychology & Marketing, 20(3), 227-248.
GfK. (2009). GfK Store Effect Survey. Niirnberg: GfK.
Gierl, H., Helm, R., & Satzinger, M. (2000). Die Wirkung positiver und negativer
Aussagen in der Werbung vor dem Hintergrund des Message Framing.
Zeitschrift far betriebswirtschaftliche Forschung, 52(3), 234-256.
Gilbert, D.T., & Hixon, J.G. (1991). The Trouble of Thinking: Activation and
Application of Stereotypic Beliefs. Journal of Personality and Social
Psychology, 60(4), 509-517.
Gilovich, T., & Medvec, V.H. (1994). The Temporal Pattern to the Experience of
Regret. Journal of Personality and Social Psychology, 67(3), 357-365.
Goldstein, M. (1997). Words to the Wise (Part 2 of2). Travel Agent, 289, 78-83.
Gourville, J.T., & Soman, D. (1998). Payment Depreciation: The Behavioral Effects of
Temporally Separating Payments from Consumption. Journal of Consumer
Research, 25(2), 160-174.
Grandey, A.A., Fisk, G.M., Mattila, A.S., Jansen, K.J., & Sideman, L.A. (2005). Is
"Service with a Smile" Enough? Authenticity of Positive Displays During
Service Encounters. Organizational Behavior and Human Decision Processes,
96(1), 38-55.
Grewal, D., Gotlieb, J., & Marmorstein, H. (1994). The Moderating Effects of
Message Framing and Source Credibility on the Price-Perceived Risk
Relationship. Journal of Consumer Research, 21(1), 145-153.
Gupta, P.B., & Gould, S.J. (1997). Consumers' Perceptions of the Ethics and
Acceptability of Product Placements in Movies: Product Category and
Individual Differences. Journal of Current Issues & Research in Advertising,
19(1), 37.
Giirhan-Canli, Z., & Maheswaran, D. (2000). Determinants of Country-of-Origin
Evaluations. Journal of Consumer Research, 27(1), 96-108.
Hahn, M., Lawson, R., & Lee, Y.G. (1992). The Effects of Time Pressure and
Information Load on Decision Quality. Psychology & Marketing, 9(5), 365-
365.
131
Hall, E. (1984). The Dance of Life. New York: Doubleday.
Heath, C., & Soll, J.B. (1996). Mental Budgeting and Consumer Decisions. Journal of
Consumer Research, 23(1), 40-52.
Herr, P.M., Kardes, F.R., & Kim, J. (1991). Effects of Word-of-Mouth and Product-
Attribute Information on Persuasion: An Accessibility-Diagnosticity
Perspective. Journal of Consumer Research, 17(4), 454-462.
Highhouse, S. (2009). Designing Experiments That Generalize. Organizational
Research Methods, 12(3), 554-566.
Ho, H.N. (2011). The Relationship between Levels of Expertise, Task Difficulty,
Perceived Self-Efficacy, and Mental Effort Investment in Task Peiformance.
Dissertation, University of Southern California, Los Angeles, CA.
Hofstede, G. (1991). Cultures and Organizations: Software of the Mind. London:
McGraw-Hill.
Hollander, S.C., & Rassuli, K.M. (1999). Shopping with Other People's Money: The
Marketing Management Implications of Surrogate-Mediated Consumer
Decision Making. Journal of Marketing, 63(2), 102-118.
Homer, P.M., & Yoon, S.-G. (1992). Message Framing and the Interrelationships
among Ad-Based Feelings, Affect, and Cognition. Journal of Advertising,
21(1), 19-33.
Hong, J., & Sternthal, B. (2010). The Effects of Consumer Prior Knowledge and
Processing Strategies on Judgments. Journal of Marketing Research, 47(2),
301-311.
Hooghe, M., Stolle, D., Maheo, V.-A., & Vissers, S. (2010). Why Can't a Student Be
More Like an Average Person?: Sampling and Attrition Effects in Social
Science Field and Laboratory Experiments. ANNALS of the American Academy
of Political and Social Science, 628(1), 85-96.
Hsee, C.K. (1995). Elastic Justification: How Tempting but Task-Irrelevant Factors
Influence Decisions. Organizational Behavior and Human Decision Processes,
62(3), 330-337.
Huang, W.-H. (2008). The Impact of Other-Customer Failure on Service Satisfaction.
International Journal of Service Industry Management, 19(4), 521-536.
Huang, W.-H. (2010). Other-Customer Failure: Effects of Perceived Employee Effort
and Compensation on Complainer and Non-Complainer Service Evaluations.
Journal of Service Management, 21(2), 191-211.
132
Iglesias-Parro, S., Ortega, A.R., De La Fuente, E.I., & Martin, I. (2001). Context
Variables as Cognitive Effort Modulators in Decision Making Using an
Alternative-Based Processing Strategy. Quality & Quantity, 35(3), 311-323.
Igou, E.R., & Bless, H. (2007). On Undesirable Consequences of Thinking: Framing
Effects as a Function of Substantive Processing. Journal of Behavioral
Decision Making, 20(2), 125-142.
Inman, J.J., & Zeelenberg, M. (2002). Regret in Repeat Purchase Versus Switching
Decisions: The Attenuating Role of Decision Justifiability. Journal of
Consumer Research, 29(1), 116-128.
ITB. (2011). ITB World Travel Trends Report 2011/2012. Berlin: Messe Berlin
GmbH.
Jaakkola, E. (2007). Purchase Decision-Making within Professional Consumer
Services: Organizational or Consumer Buying Behaviour? Marketing Theory,
7(1), 93-108.
Janis, LL., & Mann, L. (1977). Decision Making: A Psychological Analysis of
Conflict, Choice, and Commitment. New York: Free Press.
Johnson, E.J., Bellman, S., & Lohse, G.L. (2003). Cognitive Lock-in and the Power
Law of Practice. Journal of Marketing, 67(2), 62-7 5.
Johnson, E.J., & Payne, J.W. (1985). Effort and Accuracy in Choice. Management
Science, 31(4), 395-414.
Johnson, M.D. (1984). Consumer Choice Strategies for Comparing Noncomparable
Alternatives. Journal of Consumer Research, 11 (3 ), 7 41-7 53.
Jones, M.A., Mothersbaugh, D.L., & Beatty, S.E. (2002). Why Customers Stay:
Measuring the Underlying Dimensions of Services Switching Costs and
Managing Their Differential Strategic Outcomes. Journal of Business Research,
55(6), 441-450.
Kahn, B.E., & Baron, J. (1995). An Exploratory Study of Choice Rules Favored for
High-Stakes Decisions. Journal of Consumer Psychology, 4(4), 305-328.
Kahneman, D. (1973). Attention and Effort. Englewood Cliffs, New Jersey: Prentice-
Hall.
Kahneman, D., & Miller, D.T. (1986). Norm Theory: Comparing Reality to Its
Alternatives. Psychological Review, 93(2), 136-153.
Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision
under Risk. Econometrica, 47(2), 263-291.
133
Kalyuga, S., Ayres, P., Chandler, P., & Sweller, J. (2003). The Expertise Reversal
Effect. Educational Psychologist, 38(1), 23-31.
Kamakura, W.A. (2008). Cross-Selling: Offering the Right Product to the Right
Customer at the Right Time. Journal of Relationship Marketing, 6(3), 41-58.
Kanungo, R.N., & Pang, S. (1973). Effects of Human Models on Perceived Product
Quality. Journal of Applied Psychology, 57(2), 172-178.
Keaveney, S.M. (1995). Customer Switching Behavior in Service Industries: An
Exploratory Study. Journal of Marketing, 59(2), 71-82.
Khan, U., & Dhar, R. (2006). Licensing Effect in Consumer Choice. Journal of
Marketing Research, 43(2), 259-266.
Kidwell, B., & Jewell, R.D. (2008). The Influence of Past Behavior on Behavioral
Intent: An Information-Processing Explanation. Psychology & Marketing,
25(12), 1151-1166.
Kim, B.-D., & Kim, S.-0. (1999). Measuring Upselling Potential of Life Insurance
Customers: Application of a Stochastic Frontier Model. Journal of Interactive
Marketing, 13(4), 2-9.
Kim, D.-Y., Lehto, X.Y., & Morrison, A.M. (2007). Gender Differences in Online
Travel Information Search: Implications for Marketing Communications on the
Internet. Tourism Management, 28(2), 423-433.
Kim, S., & Labroo, A.A. (2011 ). From Inherent Value to Incentive Value: When and
Why Pointless Effort Enhances Consumer Preference. Journal of Consumer
Research, 38(4), 712-742.
Kimes, S.E. (2000). A Strategic Approach to Yield Management. In A. Ingold, U.
McMahon-Beattie & I. Yeoman (Eds.), Yield Management - Strategies for the
Service Industries (Vol. 2, pp. 3-14). London: South-Western.
Kirmani, A., & Zhu, R. (2007). Vigilant against Manipulation: The Effect of
Regulatory Focus on the Use of Persuasion Knowledge. Journal of Marketing
Research, 44(4), 688-701.
Klein, N.M., & Yadav, M.S. (1989). Context Effects on Effort and Accuracy in
Choice: An Enquiry into Adaptive Decision Making. Journal of Consumer
Research, 15(4), 411-421.
Kleinmuntz, D.N., & Schkade, D.A. (1993). Information Displays and Decision
Processes. Psychological Science, 4(4), 221-227.
134
Krishnamurthy, P., Carter, P., & Blair, E. (2001). Attribute Framing and Goal Framing
Effects in Health Decisions. Organizational Behavior and Human Decision
Processes, 85(2), 382-399.
Krull, D.S. (1993). Does the Grist Change the Mill? The Effect of the Perceiver's
Inferential Goal on the Process of Social Inference. Personality and Social
Psychology Bulletin, 19(3), 340-348.
Kurt, D., Inman, J.J., & Argo, J.J. (2011). The Influence of Friends on Consumer
Spending: The Role of Agency-Communion Orientation and Self-Monitoring.
Journal of Marketing Research, 48(4), 741-754.
Landman, J. (1987). Regret and Elation Following Action and Inaction - Affective
Responses to Positive Versus Negative Outcomes. Personality and Social
Psychology Bulletin, 13(4), 524-536.
Leigh, J.H., & Gabel, T.G. (1992). Symbolic Interactionism: Its Effects on Consumer
Behavior and Implications for Marketing Strategy. Journal of Consumer
Marketing, 9(1), 27-27.
Leisen, B., & Prosser, E. (2004). Customers' Perception of Expensiveness and Its
Impact on Loyalty Behaviors. Services Marketing Quarterly, 25(3), 35-52.
Lerner, J.S., & Tetlock, P.E. (1999). Accounting for the Effects of Accountability.
Psychological Bulletin, 125(2), 255-275.
Levin, LP., & Gaeth, G.J. (1988). How Consumers Are Affected by the Framing of
Attribute Information before and after Consuming the Product. Journal of
Consumer Research, 15(3), 374-378.
Levin, LP., Gaeth, G.J., Schreiber, J., & Lauriola, M. (2002). A New Look at Framing
Effects: Distribution of Effect Sizes, Individual Differences, and Independence
of Types of Effects. Organizational Behavior and Human Decision Processes,
88(1), 411-429.
Levin, LP., Schneider, S.L., & Gaeth, G.J. (1998). All Frames Are Not Created Equal:
A Typology and Critical Analysis of Framing Effects. Organizational Behavior
and Human Decision Processes, 76(2), 149-188.
Lewis, R. (2005, 20th January). The Only Way Is Up. Caterer & Hotelkeeper, 42.
Li, S., Sun, B., & Montgomery, A.L. (2011). Cross-Selling the Right Product to the
Right Customer at the Right Time. Journal of Marketing Research, 48(4), 683-
700.
Lieven, T. (2012). Female or Male Salesperson - Who "Fits" Best? Paper presented at
the 41st Annual Conference of the European Marketing Academy, Lissabon.
135
Lin, I.Y.-H., & Mattila, A.S. (2006). Understanding Restaurant Switching Behavior
from a Cultural Perspective. Journal of Hospitality & Tourism Research, 30(1),
3-15.
Liu, W. (2008). Focusing on Desirability: The Effect of Decision Interruption and
Suspension on Preferences. Journal of Consumer Research, 35(4), 640-652.
Maheswaran, D., & Meyers-Levy, J. (1990). The Influence of Message Framing and
Issue Involvement. Journal of Marketing Research, 27(3), 361-367.
Markman, A.B., & Medin, D.L. (1995). Similarity and Alignment in Choice.
Organizational Behavior and Human Decision Processes, 63(2), 117-130.
Mattila, A.S. (1999). The Role of Culture and Purchase Motivation in Service
Encounter Evaluations. Journal of Services Marketing, 13(415), 376-389.
Mellers, B.A., & McGraw, A.P. (2001). Anticipated Emotions as Guides to Choice.
Current Directions in Psychological Science, 10(6), 210-214.
Meyerowitz, B.E., & Chaiken, S. (1987). The Effect of Message Framing on Breast
Self-Examination Attitudes, Intentions, and Behavior. Journal of Personality
and Social Psychology, 52(3), 500-510.
Meyers-Levy, J., & Maheswaran, D. (2004). Exploring Message Framing Outcomes
When Systematic, Heuristic, or Both Types of Processing Occur. Journal of
Consumer Psychology, 14(112), 159-167.
Meyers-Levy, J., & Stemthal, B. (1991). Gender Differences in the Use of Message
Cues and Judgments. Journal of Marketing Research, 28(1), 84-96.
Miniard, P.W., Bhatia, S., Lord, K.R., Dickson, P.R., & Unnava, H.R. (1991). Picture-
Based Persuasion Processes and the Moderating Role of Involvement. Journal
of Consumer Research, 18(1), 92-107.
Mishra, A., & Mishra, H. (2011 ). The Influence of Price Discount Versus Bonus Pack
on the Preference for Virtue and Vice Foods. Journal of Marketing Research,
48(1), 196-206.
Mishra, S., Umesh, U.N., & Stem, D.E., Jr. (1993). Antecedents of the Attraction
Effect: An Information-Processing Approach. Journal of Marketing Research,
30(3), 331-349.
Mohr, L.A., & Bitner, M.J. (1995). The Role of Employee Effort in Satisfaction with
Service Transactions. Journal of Business Research, 32(3), 239-252.
136
Monga, A., & Houston, M.J. (2006). Fading Optimism in Products: Temporal Changes
in Expectations about Performance. Journal of Marketing Research, 43( 4), 654-
663.
Montgomery, H. (1983). Decision Rules and the Search for a Dominance Structure:
Toward a Process Model of Decision Making. In P. Humphreys, 0. Svenson &
A. Vari (Eds.), Analysing and Aiding Decision Processes (pp. 343-369).
Amsterdam: North-Holland.
Moorman, M., & van den Putte, B. (2008). The Influence of Message Framing,
Intention to Quit Smoking, and Nicotine Dependence on the Persuasiveness of
Smoking Cessation Messages. Addictive Behaviors, 33(10), 1267-1275.
Morales, A.C. (2001). It's Not Only the Thought but the Effort That Counts: The Role
of Others' Effort in Decision Making. Working Paper. The Wharton School.
University of Pennsylvania. Philadelphia, USA.
Murray, K.B., & Haubl, G. (2007). Explaining Cognitive Lock-In: The Role of Skill-
Based Habits of Use in Consumer Choice. Journal of Consumer Research,
34(1), 77-88.
Ngobo, P.V. (2005). Drivers of Upward and Downward Migration: An Empirical
Investigation among Theatregoers. International Journal of Research in
Marketing, 22(2), 183-201.
Novemsky, N., & Dhar, R. (2005). Goal Fulfillment and Goal Targets in Sequential
Choice. Journal of Consumer Research, 32(3), 396-404.
Nunes, J.C., & Boatwright, P. (2004). Incidental Prices and Their Effect on
Willingness to Pay. Journal of Marketing Research, 41(4), 457-466.
Obermiller, C., & Spangenberg, E.R. (1998). Development of a Scale to Measure
Consumer Skepticism toward Advertising. Journal of Consumer Psychology,
7(2), 159-186.
Ofek, E., Yildiz, M., & Haruvy, E. (2007). The Impact of Prior Decisions on
Subsequent Valuations in a Costly Contemplation Model. Management Science,
53(8), 1217-1233.
Okada, E.M. (2001 ). Trade-Ins, Mental Accounting, and Product Replacement
Decisions. Journal of Consumer Research, 27(4), 433-446.
Okada, E.M. (2005). Justification Effects on Consumer Choice of Hedonic and
Utilitarian Goods. Journal of Marketing Research, 42(1 ), 43-53.
Okada, E.M. (2006). Upgrades and New Purchases. Journal of Marketing, 70(4), 92-
102.
137
O'Keefe, D.J., & Jensen, J.D. (2006). The Advantages of Compliance or the
Disadvantages of Noncompliance? A Meta-Analytic Review of the Relative
Persuasive Effectiveness of Gain-Framed and Loss-Framed Messages.
Communication Yearbook, 30, 1-43.
O'Keefe, D.J., & Jensen, J.D. (2008). Do Loss-Framed Persuasive Messages Engender
Greater Message Processing Than Do Gain-Framed Messages? A Meta-
Analytic Review. Communication Studies, 59(1), 51-67.
Ord6fiez, L.D., & Connolly, T. (2000). Regret and Responsibility: A Reply to
Zeelenberg et al. (1998). Organizational Behavior and Human Decision
Processes, 81(1), 132-142.
Otto, A.R. (2010). Three Attempts to Replicate the Behavioral Sunk-Cost Effect: A
Note on Cunha and Caldieraro (2009). Cognitive Science, 34(8), 1379-1383.
Park, C.W., & Lessig, V.P. (1981). Familiarity and Its Impact on Consumer Decision
Biases and Heuristics. Journal of Consumer Research, 8(2), 223-231.
Patterson, P.G., & Smith, T. (2003). A Cross-Cultural Study of Switching Barriers and
Propensity to Stay with Service Providers. Journal of Retailing, 79(2), 107-120.
Payne, J.W. (1982). Contingent Decision Behavior. Psychological Bulletin, 92(2),
382-402.
Payne, J.W., & Bettman, J.R. (1992). Behavioral Decision Research: A Constructive
Processing Perspective. Annual Review of Psychology, 43(1), 87-131.
Payne, J.W., Johnson, E.J., Bettman, J.R., & Coupey, E. (1990). Understanding
Contigent Choice: A Computer Simulation Approach. IEEE Transactions on
Systems, Man, and Cybernetics, 20(2), 296-309.
Peterson, R.A. (2001). On the Use of College Students in Social Science Research:
Insights from a Second-Order Meta-Analysis. Journal of Consumer Research,
28(3), 450-461.
Pohlkamp, A. (2009). Identifikation und Ausschopfung von Up-Selling-Potenzialen.
Wiesbaden: Gabler.
Pompitakpan, C. (2004). The Persuasiveness of Source Credibility: A Critical Review
of Five Decades' Evidence. Journal of Applied Social Psychology, 34(2), 243-
281.
Preacher, K., & Hayes, A. (2004). SPSS and SAS Procedures for Estimating Indirect
Effects in Simple Mediation Models. Behavior Research Methods, 36(4), 717-
731.
138
Preacher, K., & Hayes, A. (2008). Asymptotic and Resampling Strategies for
Assessing and Comparing Indirect Effects in Multiple Mediator Models.
Behavior Research Methods, 40(3), 879-891.
Putrevu, S. (2010). An Examination of Consumer Responses toward Attribute- and
Goal-Framed Messages. Journal of Advertising, 39(3), 5-24.
Putrevu, S., & Lord, K.R. (1994). Comparative and Noncomparative Advertising:
Attitudinal Effects under Cognitive and Affective Involvement Conditions.
Journal of Advertising, 23(2), 77-91.
Putrevu, S., Tan, J., & Lord, K.R. (2004). Consumer Responses to Complex
Advertisements: The Moderating Role of Need for Cognition, Knowledge, and
Gender. Journal of Current Issues & Research in Advertising, 26(1), 9-24.
Raju, P.S., Lonial, S.C., & Mangold, W.G. (1995). Differential Effects of Subjective
Knowledge, Objective Knowledge, and Usage Experience on Decision Making:
An Exploratory Investigation. Journal of Consumer Psychology, 4(2), 153-180.
Rao, A.R., & Monroe, K.B. (1988). The Moderating Effect of Prior Knowledge on
Cue Utilization in Product Evaluations. Journal of Consumer Research, 15(2),
253-264.
Reh, J. (2008). Regret Aversion and Decision Process Quality: Effects of Regret
Salience on Decision Process Carefulness. Organizational Behavior and
Human Decision Processes, 105(2), 169-182.
Reh, J., & Connolly, T. (2010). The Effects of Action, Normality, and Decision
Carefulness on Anticipated Regret: Evidence for a Broad Mediating Role of
Decision Justifiability. Cognition & Emotion, 24(8), 1405-1420.
Reibstein, D.J., Youngblood, S.A., & Fromkin, H.L. (1975). Number of Choices and
Perceived Decision Freedom as a Determinant of Satisfaction and Consumer
Behavior. Journal of Applied Psychology, 60(4), 434-437.
Reinartz, W.J., & Kumar, V. (2000). On the Profitability of Long-Life Customers in a
Noncontractual Setting: An Empirical Investigation and Implications for
Marketing. Journal of Marketing, 64(4), 17-35.
Ritov, I., & Baron, J. (1990). Reluctance to Vaccinate: Omission Bias and Ambiguity.
Journal of Behavioral Decision Making, 3( 4 ), 263-277.
Roggeveen, A.L., Grewal, D., & Gotlieb, J. (2006). Does the Frame of a Comparative
Ad Moderate the Effectiveness of Extrinsic Information Cues? Journal of
Consumer Research, 33(1), 115-122.
139
Rothman, A.J., Salovey, P., Antone, C., Keough, K., & Martin, C.D. (1993). The
Influence of Message Framing on Intentions to Perform Health Behaviors.
Journal of Experimental Social Psychology, 29(5), 408-433.
Russo, J.E., & Dosher, B.A. (1983). Strategies for Multiattribute Binary Choice.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 9(4),
676-696.
Safer, D.A. (1998). Preferences for Luxurious or Reliable Products : Promotion and
Prevention Focus as Moderators. Dissertation, Columbia University, New
York.
Salazar, M.T., Harrison, T., & Ansell, J. (2007). An Approach for the Identification of
Cross-Sell and Up-Sell Opportunities Using a Financial Services Customer
Database. Journal of Financial Services Marketing, 12(2), 115-131.
Salovey, P., Schneider, T.R., & Apanovitch, A.M. (2002). Message Framing in the
Prevention and Early Detection of Illness. In J.P. Dillard & M. Pfau (Eds.), The
Persuasion Handbook: Developments in Theory and Practice (pp. 391-406).
Thousand Oaks, CA: Sage.
Samuelson, W., & Zeckhauser, R. (1988). Status Quo Bias in Decision Making.
Journal of Risk and Uncertainty, 1 ( 1 ), 7-5 9.
Schkade, D.A., & Kleinmuntz, D.N. (1994). Information Displays and Choice
Processes: Differential Effects of Organization, Form, and Sequence.
Organizational Behavior and Human Decision Processes, 57(3), 319-337.
Schmidt, J.B., & Spreng, R.A. (1996). A Proposed Model of External Consumer
Information Search. Journal of the Academy of Marketing Science, 24(3), 246.
Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D.R.
(2002). Maximizing Versus Satisficing: Happiness Is a Matter of Choice.
Journal of Personality and Social Psychology, 83(5), 1178-1197.
Selnes, F., & Troye, S.V. (1989). Buying Expertise, Information Search, and Problem
Solving. Journal of Economic Psychology, 10(3), 411-428.
Sevdalis, N., Harvey, N., & Yip, M. (2006). Regret Triggers Inaction Inertia - but
Which Regret and How? British Journal of Social Psychology, 45( 4), 839-853.
Shafir, E. (1993). Choosing Versus Rejecting: Why Some Options Are Both Better
and Worse Than Others. Memory & Cognition, 21(4), 546-556.
Shafir, E., Simonson, I., & Tversky, A. (1993). Reason-Based Choice. Cognition,
49(1-2), 11-36.
140
Sharma, A. (1990). The Persuasive Effect of Salesperson Credibility: Conceptual and
Empirical Examination. Journal of Personal Selling & Sales Management,
10(4), 71-80.
Shugan, S.M. ( 1980). The Cost of Thinking. Journal of Consumer Research, 7(2), 99-
111.
Shugan, S.M., & Xie, J. (2004). Advance Selling for Services. California Management
Review, 46(3), 37-54.
Simonson, I. (1992). The Influence of Anticipating Regret and Responsibility on
Purchase Decisions. Journal of Consumer Research, 19(1), 105-118.
Simonson, I., & Drolet, A. (2004). Anchoring Effects on Consumers' Willingness-to-
Pay and Willingness-to-Accept. Journal of Consumer Research, 31(3), 681-
690.
Slonim, R., & Garbarino, E. (1999). The Effect of Price History on Demand as
Mediated by Perceived Price Expensiveness. Journal of Business Research,
45(1), 1-14.
Smith, S.M., & Petty, R.E. (1996). Message Framing and Persuasion: A Message
Processing Analysis. Personality and Social Psychology Bulletin, 22(3), 357-
268.
Soman, D. (2001a). Effects of Payment Mechanism on Spending Behavior: The Role
of Rehearsal and Immediacy of Payments. Journal of Consumer Research,
27(4), 460-474.
Soman, D. (2001b). The Mental Accounting of Sunk Time Costs: Why Time Is Not
Like Money. Journal of Behavioral Decision Making, 14(3), 169-185.
Soman, D. (2004). The Effect of Time Delay on Multi-Attribute Choice. Journal of
Economic Psychology, 25(2), 153-175.
Soman, D., & Lam, V.M.W. (2002). The Effects of Prior Spending on Future
Spending Decisions: The Role of Acquisition Liabilities and Payments.
Marketing Letters, 13(4), 359-372.
Soster, R.L., Monga, A., & Bearden, W.O. (2010). Tracking Costs of Time and
Money: How Accounting Periods Affect Mental Accounting. Journal of
Consumer Research, 3 7( 4 ), 712-721.
Sparks, B.A., Bradley, G.L., & Callan, V.J. (1997). The Impact of Staff Empowerment
and Communication Style on Customer Evaluations: The Special Case of
Service Failure. Psychology & Marketing, 14(5), 475-475.
141
Specht, N., Fichtel, S., & Meyer, A. (2007). Perception and Attribution of Employees'
Effort and Abilities: The Impact on Customer Encounter Satisfaction.
International Journal of Service Industry Management, 18(5), 534-554.
Speier, C. (2006). The Influence of Information Presentation Formats on Complex
Task Decision-Making Performance. International Journal of Human-
Computer Studies, 64(11 ), 1115-1131.
Speier, C., Vessey, I., & Valacich, J.S. (2003). The Effects of Interruptions, Task
Complexity, and Information Presentation on Computer-Supported Decision-
Making Performance. Decision Sciences, 34(4), 771-797.
Squires, J., Wilder, D.A., Fixsen, A., Hess, E., Rost, K., Curran, R., & Zonneveld, K.
(2007). The Effects of Task Clarification, Visual Prompts, and Graphic
Feedback on Customer Greeting and Up-Selling in a Restaurant. Journal of
Organizational Behavior Management, 27(3), 1-13.
Stauss, B., & Mang, P. (1999). "Culture Shocks" in Inter-Cultural Service Encounters?
Journal of Services Marketing, 13( 415), 329-346.
Staw, B.M. (1976). Knee-Deep in the Big Muddy: A Study of Escalating Commitment
to a Chosen Course of Action. Organizational Behavior and Human
Peiformance, 16, 27-44.
Stiff, J.B., & Mongeau, P.A. (2003). Persuasive Communication (Vol. 2). New York:
The Guilford Press.
Subramanian, J., Stidham Jr., S., & Lautenbacher, C.J. (1999). Airline Yield
Management with Overbooking, Cancellations, and No-Shows. Transportation
Science, 33(2), 147-167.
Surprenant, C.F., & Solomon, M.R. (1987). Predictability and Personalization in the
Service Encounter. Journal of Marketing, 51 (2), 86-96.
Sweller, J. (1988). Cognitive Load During Problem Solving: Effects on Learning.
Cognitive Science, 12(2), 257-285.
Talluri, K., & van Ryzin, G. (2004). Revenue Management under a General Discrete
Choice Model of Consumer Behavior. Management Science, 50(1), 15-33.
Teague, L. (2011, January 8). The Outrage of the Upsell, The Wallstreet Journal, p. 6.
Tetlock, P.E. (1992). The Impact of Accountability on Judgment and Choice: Toward
a Social Contingency Model. Advances in Experimental Social Psychology, 25,
331-376.
142
Tomczak, T., Kuss, A., & Reinecke, S. (2009). Marketingplanung: Einfahrung in die
marktorientierte Unternehmens- und Geschii.ftsfeldplanung (Vol. 6).
Wiesbaden: Gabler.
Tormala, Z.L., Brifiol, P., & Petty, R.E. (2006). When Credibility Attacks: The
Reverse Impact of Source Credibility on Persuasion. Journal of Experimental
Social Psychology, 42(5), 684-691.
Trope, Y., & Liberman, N. (2000). Temporal Construal and Time-Dependent Changes
in Preference. Journal of Personality and Social Psychology, 79(6), 876-889.
Trope, Y., & Liberman, N. (2010). Construal-Level Theory of Psychological Distance.
Psychological Review, 117(2), 440-463.
Tsai, S.-P. (2007). Message Framing Strategy for Brand Communication. Journal of
Advertising Research, 47(3), 364-377.
Tuk, M.A., Verlegh, P.W.J., Smidts, A., & Wigboldus, D.H.J. (2009). Sales and
Sincerity: The Role of Relational Framing in Word-of-Mouth Marketing.
Journal of Consumer Psychology, 19(1), 38-47.
Tversky, A., & Kahneman, D. (1981). The Framing of Decisions and the Psychology
of Choice. Science, 211(4481), 453-458.
Tversky, A., & Kahneman, D. (1986). Rational Choice and the Framing of Decisions.
Journal of Business, 59(4), 251-278.
van Herpen, E., & Pieters, R. (2007). Anticipated Identification Costs: Improving
Assortment Evaluation by Diagnostic Attributes. International Journal of
Research in Marketing, 24(1), 77-88.
Vercellis, C. (2009). Business Intelligence: Data Mining and Optimization for
Decision Making. Chennai: John Wiley & Sons Ltd.
Vessey, I. (1991). Cognitive Fit: A Theory-Based Analysis of the Graphs Versus
Tables Literature. Decision Sciences, 22(2), 219-240.
von Wangenheim, F ., & Bayon, T. (2007). Behavioral Consequences of Overbooking
Service Capacity. Journal of Marketing, 71(4), 36-47.
Voss, G.B., Parasuraman, A., & Grewal, D. (1998). The Roles of Price, Performance,
and Expectations in Determining Satisfaction in Service Exchanges. Journal of
Marketing, 62(4), 46-61.
Wegener, D.T., Petty, R.E., & Klein, D.J. (1994). Effects of Mood on High
Elaboration Attitude Change: The Mediating Role of Likelihood Judgements.
European Journal of Social Psychology, 24(1), 25-43.
143
Wentzel, D. (2008). The Impact of Employee Behavior on Brand Impressions:
Theoretical and Experimental Anaylses. Dissertation, University of St. Gallen,
St. Gallen.
Wiesman, D.W. (2006). The Effects of Performance Feedback and Social
Reinforcement on Up-Selling at Fast-Food Restaurants. Journal of
Organizational Behavior Management, 26(4), 1-18.
Wiley, J. (1998). Expertise as Mental Set: The Effects of Domain Knowledge in
Creative Problem Solving. Memory & Cognition, 26(4), 716-730.
Wilkie, W.L., Mela, C.F., & Gundlach, G.T. (1998). Does "Bait and Switch" Really
Benefit Consumers? Marketing Science, 17(3), 273-282.
Wilson, D.K., Wallston, K.A., & King, J.E. (1990). Effects of Contract Framing,
Motivation to Quit, and Self-Efficacy on Smoking Reduction. Journal of
Applied Social Psychology, 20(7), 531-547.
Winer, R. (1999). Experimentation in the 21st Century: The Importance of External
Validity. Journal of the Academy of Marketing Science, 27(3), 349-358.
Wirtz, J., Kimes, S.E., Ho, J., & Patterson, P.G. (2003). Revenue Management:
Resolving Potential Customer Conflicts. Journal of Revenue and Pricing
Management, 2(3), 216-226.
Witkowski, T.H., & Wolfinbarger, M.F. (2002). Comparative Service Quality:
German and American Ratings across Service Settings. Journal of Business
Research, 55(11), 875-881.
Woods, C. (2006). Careless Responding to Reverse-Worded Items: Implications for
Confirmatory Factor Analysis. Journal of Psychopathology and Behavioral
Assessment, 28(3), 186-191.
Wooten, D.B., & Reed II, A. (2004). Playing It Safe: Susceptibility to Normative
Influence and Protective Self-Presentation. Journal of Consumer Research,
31(3), 551-556.
World Trade Organization. (2011 ). Trade in Commercial Services. In World Trade
Organization (Ed.), International Trade Statistics 2011 (pp. 134-182). Geneva,
Switzerland: WTO.
Yeo, G., & Neal, A. (2008). Subjective Cognitive Effort: A Model of States, Traits,
and Time. Journal of Applied Psychology, 93(3), 617-631.
Yeung, C.W.M., & Soman, D. (2007). The Duration Heuristic. Journal of Consumer
Research, 34(3), 315-326.
144
Yi, S., & Baumgartner, H. (2008). Motivational Compatibility and the Role of
Anticipated Feelings in Positively Valenced Persuasive Message Framing.
Psychology & Marketing, 25(11), 1007-1026.
Zauberman, G. (2003). The lntertemporal Dynamics of Consumer Lock-In. Journal of
Consumer Research, 30(3), 405-419.
Zeelenberg, M. (1999). Anticipated Regret, Expected Feedback and Behavioral
Decision Making. Journal of Behavioral Decision Making, 12(2), 93-93.
Zeelenberg, M., & Pieters, R. (2007). A Theory of Regret Regulation 1.0. Journal of
Consumer Psychology, 17(1), 3-18.
Zeelenberg, M., van den Bos, K., van Dijk, E., & Pieters, R. (2002). The Inaction
Effect in the Psychology of Regret. Journal of Personality and Social
Psychology, 82(3), 314-327.
Zeelenberg, M., van der Pligt, J., & de Vries, N.K. (2000a). Attributions of
Responsibility and Affective Reactions to Decision Outcomes. Acta
Psychologica, 104(3), 303-315.
Zeelenberg, M., van Dijk, W.W., & Manstead, A.S.R. (1998). Reconsidering the
Relation between Regret and Responsibility. Organizational Behavior and
Human Decision Processes, 74(3), 254-272.
Zeelenberg, M., van Dijk, W.W., Manstead, A.S.R., & van der Pligt, J. (2000b). On
Bad Decisions and Disconfirmed Expectancies: The Psychology of Regret and
Disappointment. Cognition & Emotion, 14(4), 521-541.
Zhang, J., Beatty, S.E., & Walsh, G. (2008). Review and Future Directions of Cross-
Cultural Consumer Services Research. Journal of Business Research, 61(3),
211-224.
Zhang, S., & Fitzsimons, G.J. (1999). Choice-Process Satisfaction: The Influence of
Attribute Alignability and Option Limitation. Organizational Behavior and
Human Decision Processes, 77(3), 192-214.
Zhang, S., & Markman, A.B. (1998). Overcoming the Early Entrant Advantage: The
Role of Alignable and Nonalignable Differences. Journal of Marketing
Research, 35(4), 413-426.
Zhao, X., Lynch, J.G.J., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths
and Truths about Mediation Analysis. Journal of Consumer Research, 3 7(2),
197-206.
Zhou, R., & Soman, D. (2003). Looking Back: Exploring the Psychology of Queuing
and the Effect of the Number of People Behind. Journal of Consumer Research,
29(4), 517-530.
145
7 Appendices
7 .1 Appendix 1: Stimulus Materials used in Experiment 1
a. Introductory Ieformation (in all Conditions)
Soon it will happen - the semester break is approaching! Together with your friends
you plan to travel to Berlin for a long weekend to enjoy the city flair, go shopping and
to party. You have already booked your flight from Zurich to Berlin. For the first two
nights and days you will be on your own since your friends will still be working on
their last exams. Therefore, you need a hotel room for these nights.
Consequently, you visit the homepage of a booking portal and enter the stay of two
nights as well as your price limit of 50 Euros per night. Finally, you choose a three-star
hotel ("Hotel Charlott") in the heart of Berlin, from where you can reach most of the
attractions within walking distance. Hence, all you have to do is to decide for one of
the room categories. The following rooms are available in the selected hotel (2
nights/50 Euros per night):
b. Low Cognitive Effort Manipulation
IM =J?:··
litli\lilitl~~tlli\llll~~~~~~~:i:\i~~\~i~~~~~lllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll~llllllllllllllllllllllllllllllllllllllllll
::::::::::::::::::
... in the Heart .of Berlin
mn r • 6 rrttnr• f nt'b er m·mme•
146
c. High Cognitive Effort Manipulation
Hotel Charlott***
?' m tt+ ,., + t
••• in the Haart of Berlin
' • 7 + ,,, $ b umf!m; ,. t r ·m m m:
Dti~• of Arrivai > Available RooMi > Re~t1tio11
~rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr~~
<:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·
RomnB:
Ifoorn 0:
rhe rnt>m e1i.t~0mp;i;;se;. 24 sq m !t IJ;;;; one l>.-';;:11 ;!~• a b::1thtuh. ·For br~akfast yon can s;;:rv~ y<)Cff~•·e1f
at the huifot
Thx". .rootn '2t~:i:0tnp;bs~:~ 30 sq fft and is frn.1·1~~hed '\.~:ith .,:)ne Jt)~d. You h~1.V(: 3nh:~fll\:'!t acc:\:·~s ·vfa \\:·~Fi .~\-<.idit101·!il11y~
ih~ n .. X)lH is equipped 'l;~·ith a safe .. ·~ ~::.tiniba:.::- a \\Thing desk a~ ~~.--ell as :i hair:st s.ho\ver . .for breS:kfa~t you catJ ~i!rvi:: y~_'l"nrsc:ff~; 1he bnffot \Jr use ;he r<)on1 servif:e.
Tht: ro<.~xn-encornpas::~s i:' sq rn. 'ft has nv~~ be.d:~. au./ff} i\d<.1it~onaUy~ th~ roo.1n is eql1ipp\:X1 \Vjib. a Si!ff.:\
s n1in.ibar :i.~· ~i-·ell as a h1~frdry·er. Yen al~o h~ive rh~ wake·-HP .sen·icc at y~~ur dhp·~~:-+at In ihe b;}ih.n:n:nn you cau
fiw"i :.5.B i~rcrn;::, s!K1~:-..;·c;".J~ f\x br~a:kfi:i:St yo:o..1 t~an serve ·your~elf at th~ buffet or !..~::it the r>:~'t1.Jn1 ~t:rvk'.e.
d. Introduction Goal Framing (in all Conditions)
The time has come. The semester break has just begun and you disembarked in Berlin.
First, you set out for your hotel in order to check in. When you finally arrive at the
hotel you give your name and booking number to the desk clerk. While the
receptionist checks your reservation details she shows interest in your journey and the
reason for your stay in the capital. You tell her about the long weekend with your
friends and what you wish to see and experience. Meanwhile she has called up your
booking details, checks them and says:
147
e. Gain-Framed Upsell Offer
"I see that a room became available this morning that I can offer you instead of the
room you had reserved. It would only cost you 10 Euros more per night. It has two
advantages:
First, its windows face the river. In the morning you will have a beautiful view toward
the sunset over the River Spree. The room is back to the street and its background
noise, so you can have a restful sleep after an exhausting day in the city. On the next
day you will be fit and relaxed.
Moreover, this room also encompasses a king-sized bed. Although you are traveling
on your own, you can relax and spread yourself out on a 1.5m x 2.0m mattress. I see
that you travel with a lot of luggage. When you scatter your belongings on the bed,
there will still be enough space left for you. So enjoy relaxed days and decide in favor
of this offer. You will be absolutely satisfied."
f Loss-Framed Upsell Offer
"I see that a room became available this morning that I can offer you instead of the
room you had reserved. It would only cost you 10 Euros more per night. The room you
had reserved has two disadvantages compared to my offer:
First, its windows face the street. In the morning you will miss the sunset over the
River Spree. If you are not used to the permanent background noise of the street, it
might trouble you. After an exhausting day in the city, there might be nothing worse
than feeling tired the next day. Instead, the room that I can offer you would face the
quiet riverside.
Moreover, the room you had reserved only encompasses a standard sized bed for one
person. You cannot spread yourself out on the bed. Although you travel on your own,
you carry lots of luggage with you, as I can see. When you scatter your belongings on
the bed, there will not be enough space left for you. Relaxation might become difficult.
The room that I can offer you has a king-sized bed with a 1.5m x 2.0m mattress. So,
beware of making a mistake and grab this offer. You will not regret it."
148
7.2 Appendix 2: Stimulus Materials used in Experiment 2
a. Introductory Ieformation (in all Conditions)
Soon it will happen - your next vacation is approaching! In this year's autumn
holidays of one week, you want to discover Germany and meet some friends on the
way. After beginning your road trip in Kiel, you want to rent a car to drive all the way
to Munich. You have already booked a flight to Kiel some time ago. So the only thing
you have to do is reserve a rental car, with which you can drive from Kiel to Munich.
You therefore visit the homepage of the car rental company "RentThisCar" and enter
the rental period of one week as well as your price limit of 50 Euros per day. Hence,
all you have to do is decide which car to reserve. The screenshot of the following page
depicts the options that are available to you.
b. Low Cognitive Effort Manipulation
149
c. High Cognitive Effort Manipulation
ii mm:r························
liTuisl\ll~iliil:~~(::i~~~::~~:~~:~::~l~~:~~~::~E~:~L~~~~\~:~:~~E~&~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::111:::::::::::::::::::::::::::::::::::::::1
This rnrfealuresltirnedoors. The trunk !ms spllce for one large suitcase. The carnffe(s9Gllp and Mas an
automatic trarismission. The cam capable of do:ng1CO~m on 5 S lite:s otl'-.1el. tt is addi!!onally eqwppeel wtth
two a;rbags and air condillomr.g.
Th!s car feature~, four doors a11C; the lrur.k nas sp~;ce fm two smal! suitcases. Ti1e car offers 75 hp and i1as a
manual transmission. The car !s capacle of doir.~ ·n)Okm on 5.5 liiers ol:foet !l is adrJitiona!tt equipped witll 1wo
aimags and a :11ult:-me1'iit'\ s~·ste•)1
This car fea!JJres tour doors aM i$ equipped with two airbags and a;r comJ;t1cming The tru11k has space tor one
large ar.d one sn1al! suilca;se. The car offers 60 l'lp and has an aulomatic trnn:>m!ssion ll is c.3pabh> of doing
iOOkm cm 5.5 ltters of iueL
This car features U;ree doors and the trnnk has si1ace !or one large ancl one sma!l sl!ltcase. Tue ;.oar offers 75 hp
aml tias a manual iransm~ss;mL it is capable ofooing 100km on 5Jl!!tern offuet it is addi!icmally equipped with
t'~{O a;rpa~1s.
d. Introduction to the Film (in all Conditions)
Prior to seeing the video, each study participant was subjected to the following
introduction:
A few weeks later, the time has come: Your vacation has just begun and you have
arrived in Kiel. After you disembarked the plane you picked up your suitcases and
went to the rental office where you had reserved the car for your journey.
On the next page, you will find a video that depicts the interaction between you and
the representative of the car rental company. Although you might have reacted
differently than the customer shown in the video - imagine you had experienced the
situation in his stead.
150
e. Video Vignette Gain-Framed Upsell Offer
C= Customer, E= Employee of the Car Rental Company
[Fade-in: The representative of the car rental company is working on his computer.
The customer enters the scene and steps to the counter.}
C: Good morning!
E: [looks up from the screen} Good morning! How can I help you?
C: I have a reservation for a car and would like to pick it up.
E: So I need your reservation number.
C: Yeah, I think the information is on this confirmation here. [hands the reservation
confirmation over to the counter employee}
E: OK, I will check this in the system. [enters the reservation number in the computer
and reads the details on the screen} Oh, I can see you are driving to Munich?
C: Yes, that is right. I am on vacation for one week and want to meet some friends on
the way.
E: And the baggage is yours? [points to a place behind the customer}
C: Yes, this is mine.
E: I am asking for the following reason. I just got a car from a higher vehicle type back
that I can offer you for 10 Euros more per day.
C:Hmmm.
E: This car has two clear advantages compared to the car you had reserved.
C:Oh?
E: First, it is bigger and offers you more comfort. You can stow your luggage
completely in the trunk. The passenger cabin remains free for additional souvenirs,
friends and acquaintances.
C: Hmmm, and are there more differences compared to my reserved car?
E: Yes, the car I can offer you has an integrated cruise control. With this device you
can fix the tempo. Especially on this long journey, this allows you to take your foot
off the accelerator. Your feet remain relaxed; you will need fewer stops and will
probably arrive earlier at your destination.
C: OK, I understand this.
E: So enjoy a relaxed car ride and grab this offer. You will be satisfied.
[Fade-out]
151
f Video Vignette Loss-Framed Upsell Offer
C= Customer, E= Employee of the Car Rental Company
[Fade-in: The representative of the car rental is working on his computer. The
customer enters the scene and steps to the counter.}
C: Good morning!
E: [looks up from the screen} Good morning! How can I help you?
C: I have a reservation for a car and would like to pick it up.
E: So I need your reservation number.
C: Yeah, I think the information is on this confirmation here. [hands the reservation
confirmation over to the counter employee}
E: OK, I will check this in the system. [enters the reservation number in the computer
and reads the details on the screen} Oh, I can see you are driving to Munich?
C: Yes, this is right. I am on vacation for one week and want to meet some friends on
the way.
E: And the baggage is yours? [points to a place behind the customer}
C: Yes, this is mine.
E: I am asking for the following reason. I just got a car from a higher vehicle type back
that I can offer you for 10 Euros more per day.
C:Hmmm.
E: The car you had reserved has two clear disadvantages compared to my offer.
C:Oh?
E: The car you had reserved is smaller and offers you less comfort. The trunk is too
small to stow your luggage completely. So you'll have to take some pieces with you
in the passenger cabin. Therefore, it might become difficult to take friends,
acquaintances or additional souvenirs with you.
C: Hmmm, and are there any more differences compared to your offer?
E: Yes, the car I can offer you boasts an integrated cruise control. With this device you
can fix the tempo. The car you had reserved is not equipped with this device.
Considering the length of your journey, this implies that your foot must remain on
the accelerator. Thus, your feet are likely to get weary more quickly; you will need
more stops and will probably arrive later at your destination.
C: OK, I understand this.
E: So spare yourself an exhausting car ride and grab this offer. You will not regret it.
[Fade-out]
152
7.3 Appendix 3: Stimulus Materials used in Experiment 3 (Car Rental)
a. Introductory Ieformation (in all Conditions)
Soon it will happen - the semester break is approaching! In this year's summer break
for one week, you want to discover Germany and meet some friends and fellow
students on the way. After beginning your road trip in Kiel, you want to rent a car to
drive all the way to Munich (ca. 900 km). You have already booked a flight to Kiel
some time ago. The only thing you have to do is to reserve a rental car, with which you
can drive from Kiel to Munich.
Consequently, you visit the homepage of the car rental company "RentThisCar" and
enter the rental period of one week as well as your price limit of 50 Euros per day.
Hence, all you have to do is to decide which car to reserve. The following screenshot
depicts all options that are available to you.
b. High Cognitive Effort Manipulation (in all Conditions)
Tn~S! ca:: ·f~ot\ires th>a.;1 tiaors. The trur:k ~as Spaye for· one :al'ye s~ik:iise-: The· £ar otre~ BO hp ar.d ha:s ar~ at!tcmati·C:
tr~rt~mi5sk)r:. The car :$.capt3.bfe cf d:1~r:g 1CCkm or> :$ . .E me:-s C:f fuaL !t :s adtf:tior.a3y atJD~pp~ \'-:lt~·k ~~·t't ... "S~ffi39s. gnd ai:-
.:::cnO~t~i~ntr:g
T:~is c»r f2~rures fo:.:r doors ~nd th-e tmnk ha:o. sp-ac~ fo~·t"~'o sr~&!l. sul.t.~ase~. T~~~· car offe~-s 75 hp .and has a mar.u&!
rransm$:;.sl:On. The :;a; ;:s t:apabl€ of dotng ~O.Qkrn on 5,5 laB:-s :0ffi~~et H :s· a aM}ags 'and a rnul:;-
ma:d:a: sys~;:;m.
Th;B car fea~ure~ fo:...:.r tloDrn anQ ~s €q1sipped wkh t'Nc ~irbags ami a:r 00r.dit~0::hi:g lhe i~u:~~: has s~ce fo~ one iar~e arid
o~e small suRcgse. Th.e car c:ffers 60 h:) and ~as. at1 avtcmatfc transmjs~m:m. ii :.se:ap~o:e of fJ:o~n~ 1:CDkm en 5.S :~t2rs of
fc;i;I.
This <:ia~ fo.::tureS t~w~~ dc<1rs ·~nct the tl'i.Jt:k h~~ spa<·~· for or:~ ia;ge ·an~:J >)ne:srnall *u:tca~~, The car offers 7:6 hp an.~ has ~
!"!1ar:t.::ia: tra~sm~~s~on. tt is e;~pab;~ oJ {k,~;tig· ·J'OCkrn o:n 5.8 llt~r~ Df fu:el. lt :s ad.::'Htlo!"laHy' &quipped w~i:h h.'io afrt~gs,
Thj:s. t·a:»feak.~res. f::>:.:r rj.;JOf% ar:d·1$ eG:..1·fpp:ed ;.r...1lth ti.vo a;rbags:at•d a:r c~::i:~ct:ttontn-;;. The lrunk 1v.ls sµa~::e tor tw:) sma!~
:>uttt as·es. The .::.(l}~ c-ff'.ers 79 h:p. h~~ a m;sr!ua! trants.n)::;~~on an-q ~s t.ilpabl>!-·l)f d¢~f:g 1 OCf.im ·or: .5 .a }ite~~ ~-:.f h:e(
153
c. Control Condition & Gain-Framed Upsell Offer
The time has come. The semester break has just begun and you have arrived in Kiel.
After you disembarked and collected your luggage you proceeded to the car rental
company where you had reserved your car. When you finally arrive at the car rental
office you give your name and booking number to the employee. The car rental
representative calls up your reservation details, checks them and says to you:
"I see that a car from a higher category became available this morning that I can offer
you instead of the car you had reserved. It would only cost you 10 Euros more per day.
This offer has two clear advantages compared to the car you had reserved.
First, it is bigger and offers you more comfort. I see that you travel with a lot of
luggage. The trunk of this car is big enough to stow your luggage completely. The
passenger cabin remains free for friends and acquaintances or bigger souvenirs.
Moreover, the car I can offer you comprises an integrated cruise control. With this
device you can fix the tempo. Especially on this long journey, this allows you to take
your foot off the accelerator. Your feet remain relaxed; you will need fewer stops and
will arrive earlier at your destination. So enjoy a relaxed car ride and grab this offer.
You will be satisfied."
d. Control Condition & Loss-Framed Upsell Offer
The time has come. The semester break has just begun and you have arrived in Kiel.
After you disembarked and collected your luggage you proceeded to the car rental
company where you had reserved your car. When you finally arrive at the car rental
office you give your name and booking number to the employee. The car rental
representative calls up your reservation details, checks them and says to you:
"I see that a car from a higher category became available this morning that I can offer
you instead of the car you had reserved. It would only cost you 10 Euros more per day.
The car you had reserved has two clear disadvantages compared to my offer.
The car you had reserved is smaller and offers you less comfort. I see that you travel
with a lot of luggage. The trunk of the reserved car is too small to stow your luggage
completely. So you'll have to take some pieces with you in the passenger cabin.
Therefore, there might not be ample space to take friends, acquaintances or bigger
souvenirs with you. Moreover, the car you had reserved is not equipped with a cruise
control. Accordingly, you can't fix the tempo. Considering the length of your journey,
this implies that your foot must remain on the accelerator. Thus, your feet are likely to
154
get weary more quickly; you will need more stops and will arnve later at your
destination. The car I can offer you has no such disadvantages. So spare yourself an
exhausting car ride and grab this offer. You will not regret it."
e. No-Justification Condition & Gain-Framed Upsell Offer
The time has come. The semester break has just begun. Unfortunately, your flight to
Kiel departed with some hours delay. Therefore, you have asked a good friend from
Kiel to pick up the reserved car and to advance the rental costs so you can refund him
the costs afterwards. Your friend sets off to the car rental company where you had
reserved your car and gives your name and reservation number to the car rental
representative. The representative calls up your reservation details, checks them and
says to your friend:
"I see that a car from a higher category became available this morning that I can offer
you instead of the reserved car. It would only cost 10 Euros more per day. This offer
has two clear advantages compared to the reserved car.
First, it is bigger and offers more comfort. Usually one needs many bags for such trips.
The trunk of this car is big enough to stow lots of luggage completely. The passenger
cabin remains free for friends and acquaintances or bigger souvenirs. Moreover, the
car I can offer you comprises an integrated cruise control. With this device you can fix
the tempo. Especially on this long journey, this allows you to take your foot from the
accelerator. Your feet remain relaxed; you will need fewer stops and will arrive earlier
at your destination. So enjoy a relaxed car ride and grab this offer. You will be
satisfied."
f No-Justification Condition & Loss-Framed Upsell Offer
The time has come. The semester break has just begun. Unfortunately, your flight to
Kiel departed with some hours delay. Therefore, you have asked a good friend from
Kiel to pick up the reserved car and to advance the rental costs so you can refund him
the costs afterwards. Your friend sets off to the car rental company where you had
reserved your car and gives your name and reservation number to the car rental
representative. The representative calls up your reservation details, checks them and
says to your friend:
155
"I see that a car from a higher category became available this morning that I can offer
you instead of the reserved car. It would only cost 10 Euros more per day. The
reserved car has two clear disadvantages compared to my offer.
The reserved car is smaller and offers less comfort. Usually one needs many bags for
such trips. The trunk of the reserved car is too small to stow lots of luggage
completely. Some pieces will have to be taken in the passenger cabin. Therefore, there
might not be ample space to take friends, acquaintances or bigger souvenirs along.
Moreover, the reserved car is not equipped with a cruise control. Accordingly, you
can't fix the tempo. Considering the length of this journey, this implies that your foot
must remain on the accelerator. Thus, your feet are likely to get weary more quickly;
you will need more stops and will arrive later at your destination. The car I can offer
you has no such disadvantages. So spare yourself an exhausting car ride and grab this
offer. You will not regret it."
7.4 Appendix 4: Stimulus Materials used in Experiment 3 (Hotel)
a. Introductory Ieformation (in all Conditions)
Soon it will happen - the semester break is approaching! Together with your friends
you plan to travel to Berlin for a long weekend to enjoy the city flair, go shopping and
to party. You have already booked your flight from Zurich to Berlin some time ago.
For the first two nights and days you will be on your own since your friends will still
be working on their last exams. Therefore, you need a hotel room for these nights.
Consequently, you visit the homepage of a booking portal and enter the stay of two
nights as well as your price limit of 50 Euros per night. Finally, you choose a three-star
hotel ("Hotel Charlott") in the heart of Berlin, from where you can reach most of the
attractions within walking distance. Hence, all you have to do is to decide on one of
the rooms. The following rooms are available in the selected hotel (2 nights/50 Euros
per night):
156
b. High Cognitive Effort Manipulation (in all Conditions)
Day ot Arrival $ Au~dl1:1ble, ffOOllU :Jt Res;;>;vatlon
·~~~.,-•. ~"'"''~•························-• The room er:compas$e::~ -241!-q m. ff ha$ or:e ~ct and cffers VViF\ .6r.dd~W:ooal~}', the. room f$ equ~p;>ed \'\'ifu a sate., a mk~·th~~
~~ -..~·~"!: as a h,atrdryer, ;r: the: batr)rc~c:lm :fo.v car) ttnc: a ra~nfcre.stshov.~r. F°i:){ Oreakf:sstyou tan s~~v~ yo·rn:"se:t at m= !x:ffet •3!'
use th:s ri:,,.,:,~m se-r,)ic~.
The ~ccm ~-nc::>m~s.ses:_.20 sq m. U ha·$ ~1,~·o bet!s. and Qff*t~ VVtPt, A', the ~ccm is eqi.i!pped with~: ~ate arj1j a
f":;ii:ctrye;r. !>'l ~t:e !li.11!1n:iu111 you ca~ fi~d a s~iower >l'> weli as ;i batni\:b, F<:Jc !:::ower.
fot b~.;Jki'ast yoU c~r: sen;e :;.-·ou:'s&tf at the· buffet or ~:se the r:.':>oi1) s~:vtce,
c. Control Condition & Gain-Framed Upsell Offer
The time has come. The semester break has just begun and you disembarked in Berlin.
First, you set out for your hotel in order to check in. When you arrive at the hotel you
give your name and booking number to the desk clerk. The receptionist calls up your
reservation details, checks them and says to you:
"I see that a room from a higher category became available this morning that I can
offer you instead of the room you had reserved. It would only cost 10 Euros more per
night. This offer has two clear advantages compared to the room you had reserved.
This room faces the quiet riverside. In the morning you will have a beautiful view
toward the sunset over the river. On vacation one usually wants to sleep late. The room
that I can offer you is quietly located and enables you to have a peaceful night after an
exhausting day in the city. There is nothing better than starting relaxed the next day.
Moreover, instead of a standard-sized bed for one person this room encompasses a
larger king-sized bed. Although you are traveling on your own, I can see that you
travel with a lot of luggage. The bed in the room that I can offer you is broad enough
so you can spread your belongings on it with still enough space left for you. So enjoy
relaxed days and decide in favor of this offer. You will be absolutely satisfied."
157
d. Control Condition & Loss-Framed Upsell Offer
The time has come. The semester break has just begun and you disembarked in Berlin.
First, you set out for your hotel in order to check in. When you arrive at the hotel you
give your name and booking number to the desk clerk. The receptionist calls up your
reservation details, checks them and says to you:
"I see that a room from a higher category became available this morning that I can
offer you instead of the room you had reserved. It would only cost 10 Euros more per
night. The room you had reserved has two clear disadvantages compared to my offer.
The room you had reserved faces the noisy street side. In the morning you will miss
the sunset over the river. On vacation one usually wants to sleep late. In the room you
had reserved you might be troubled by the background noise of the street in the
morning. After an exhausting day in the city there is nothing worse than starting the
next day tiredly. Moreover, instead of a king-sized bed the room you had reserved only
encompasses a smaller standard-sized bed for one person. Although you are traveling
on your own, I can see that you travel with a lot of luggage. The bed in your reserved
room is too confined to spread your belongings on it. Otherwise, not enough space will
be left for you. The room that I can offer you does not suffer these disadvantages. So
beware of a stressful stay and decide in favor of this offer. You will not regret it."
e. No-Justification Condition & Gain-Framed Upsell Offer
The time has come. The semester break has just begun. Unfortunately, your flight to
Berlin departed with some hours delay. Since the reception in the hotel in Berlin is
closing soon you have asked a friend from Berlin to check in for you and to advance
the room rate so you can refund him the costs afterwards. Your friend sets off to the
hotel und gives your name and booking number to the desk clerk. The receptionist
calls up your reservation details, checks them and says to your friend:
"I see that a room from a higher category became available this morning that I can
offer you instead of the reserved room. It would only cost 10 Euros more per night.
This offer has two clear advantages compared to the reserved room.
This room faces the quiet riverside. In the morning you will have a beautiful view
toward the sunset over the river. On vacation one usually wants to sleep late. The room
that I can offer you is quietly located and enables you to have a peaceful night after an
exhausting day in the city. There is nothing better than starting relaxed the next day.
Moreover, instead of a standard-sized bed for one person this room encompasses a
158
larger king-sized bed. Although traveling on your own, one usually travels with a lot
of luggage while on city trips. The bed in the room that I can offer you is broad
enough so you can spread your belongings on it with still enough space left for you. So
enjoy relaxed days and decide in favor of this offer. You will be absolutely satisfied."
f No-Justification Condition & Loss-Framed Upsell Offer
The time has come. The semester break has just begun. Unfortunately, your flight to
Berlin departed with some hours delay. Since the reception in the hotel in Berlin is
closing soon you have asked a friend from Berlin to check in for you and to advance
the room rate so you can refund him the costs afterwards. Your friend sets off to the
hotel und gives your name and booking number to the desk clerk. The receptionist
calls up your reservation details, checks them and says to your friend:
"I see that a room from a higher category became available this morning that I can
offer you instead of the reserved room. It would only cost 10 Euros more per night.
The reserved room has two clear disadvantages compared to my offer.
The reserved room faces the noisy street side. In the morning you will miss the sunset
over the river. On vacation one usually wants to sleep late. In the reserved room you
might get troubled by the background noise of the street in the morning. After an
exhausting day in the city there is nothing worse than starting the next day tiredly.
Moreover, instead of a king-sized bed the reserved room only encompasses a smaller
standard-sized bed for one person. Although traveling on your own, one usually travels
with a lot of luggage while on city trips. The bed in the reserved room is too confined
to spread your belongings on it. Otherwise, not enough space will be left for you. The
room that I can offer you does not suffer these disadvantages. So beware of a stressful
stay and decide in favor of this offer. You will not regret it."
159
7.5 Appendix 5: Stimulus Materials used in Experiment 4
a. Introductory Ieformation (in all Conditions)
Imagine: The end of term is approaching. To reward yourself for the hard work of the
last months, you plan a one-week trip to Germany to meet some of your friends. You
want to start your trip in Berlin, rent a car, visit some cities and then drive back to
St. Gallen. In order to book the flight and to reserve a rental car, you visit a nearby
travel agency and tell the agent of your travel plans. Prior to this, you have already
selected a flight that only needs to be booked now. But when it comes to choosing and
reserving the rental car you need the assistance and expertise of the travel agent. You
have set yourself a price limit of 80 Swiss Francs [58 Euros] per rental day. The travel
agent notes this information, browses for a proper offer in his system and says:
b. Low Perceived Effort Condition
"Let me see what we have. The variety of available cars is really comprehensive. Do
you have a clear idea of the car you want to rent for a week?"
You reply: "To be honest, I am not experienced in renting cars. I am sure you have
done that a lot of times - I leave the reservation to your discretion."
Thereupon, the travel agent shortly considers the options and says: "After all the things
you told me about your trip, I can quickly tell which of the available cars is the best
deal for you. I would reserve this car here [he points on his computer screen} - I think
you will like it. Besides, it is in your price range of 80 Swiss Francs a day. When you
agree, just give me your signature and I will reserve the car for you."
You look at the reservation contract in front of you, sign it and say: "That was fast.
Now my holidays are ready to start!"
You pack up your belongings, say goodbye and leave the travel agency.
160
b. High Perceived Effort Condition
"Let me see what we have. The variety of available cars is really comprehensive. Do
you have a clear idea of the car you want to rent for a week?"
You reply: "To be honest, I am not experienced in renting cars. I am sure you have
done that a lot of times - I leave the reservation to your discretion."
Thereupon, the travel agent buries himself in the reservation decision for some
minutes and steadily informs you about the criteria he uses to select the appropriate
car. Finally, he says to you: "I have narrowed the extensive range of available offers to
three cars. But let me peruse the exact car descriptions before I finally reserve one of
them." Again, he reads through all the offers, compares them and says: "Well, I think,
after all the things you told me about your trip, this car here [he points on his computer
screen} is most appropriate for you and it is the best deal. Besides, it is in your price
range of 80 Swiss Francs a day. When you agree, just give me your signature and I
will reserve the car for you."
You look at the reservation contract in front of you, sign it and say: "Thank you for
your advice and extensive assistance in choosing a rental car. Now my holidays are
ready to start!"
You pack up your belongings, say goodbye and leave the travel agency.
c. Introduction Goal Framing (in all Conditions)
The time has come. The semester break has just begun and you disembarked in Berlin.
Directly after landing you proceed to the counter of the car rental company where your
car was reserved. To hit the road you only have to sign the rental contract. You give
your reservation number to the representative of the car rental company. While the
computer searches for your reservation information, the representative asks you about
your destination. You tell him about your planned road trip through Germany and how
much you are looking forward to it. Thereupon, the representative checks the
reservation details and says:
161
d. Gain-Framed Upsell Offer
"I see that a car from a higher category just became available that I can offer you for a
small premium of 18 Swiss Francs [13 Euros] per day.
The car is significantly bigger than the one you had reserved and has more extra
equipment. When you choose this car instead of the one you had reserved you will do
the right thing. I can guarantee you will have memorable holidays. Considering the
length of your journey, the car I can offer you is very comfortable -you can drive long
distances without thinking about it. Due to the bigger size, this car offers you driving
enjoyment and more power.
The cabin space of the car that I can offer you is spacious enough to travel with friends
and enjoy the vacation together with them. I see that you travel with a lot of luggage.
There is enough space to stow it in the car while some space will additionally be left
for souvenirs. So enjoy a relaxed vacation and grab this offer. I cannot guarantee the
availability later on. With this car you receive top performance for your money. You
will be satisfied."
e. Loss-Framed Upsell Offer
"I see that a car from a higher category just became available that I can offer you for a
small premium of 18 Swiss Francs [13 Euros] per day.
The car you had initially reserved is significantly smaller and has less extra equipment.
When you stick to your reservation instead of taking this offer you might become
worried. Considering the length of your journey, the reserved car is not as comfortable
- driving long distances might become inconvenient. Due to the small size, the
reserved car has lower power. Driving enjoyment might deteriorate.
I see that you travel with a lot of luggage. The cabin space of the reserved car is
limited, so you will surely encounter space problems. Especially when you plan to
travel with friends you might have to leave some pieces. The car that I can offer you
has no such disadvantages. So beware of stressful holidays and grab this offer. I cannot
guarantee the availability later on. So get more out of you holidays and your money.
You will not regret it."

Curriculum Vitae
Name
Date of Birth
Education
2009-2012
2006-2008
2002-2005
1994-2002
Work Experience
2008-2013
2007-2008
2006-2007
2002-2005
Wibke Heidig
May 19, 1983
University of St.Gallen, Switzerland
Doctoral Studies in Business Administration
University of Michigan, Ann Arbor, USA
Summer Program in Quantitative Methods of Social Research
Technical University of Berlin, Germany
Diploma Studies in Business Administration
Tongji University, Shanghai, China
Exchange Studies
University of Cooperative Education Ravensburg, Germany
Bachelor Studies in Business Administration
Katharina-von-Hagenow Gymnasium Barth, Germany
Abitur
Center for Customer Insight, University of St.Gallen, Switzerland
Research Associate
Chair of Marketing, Technical University of Berlin, Germany
Tutor & Teaching Assistant
Chair of Human Resource Management and Intercultural
Leadership, ESCP Europe Campus Berlin, Germany
Student Assistant
KarstadtQuelle AG & Madeleine Mode GmbH, Fiirth, Germany
Trainee in Marketing, Human Resources, Controlling,
Public Relations, Purchasing, and Quality Control
Why cross .. seUing and upseUing seem so difficult
to implement
Telemarketing & Call Center Solutions
January 1, 1996 I Levine, Larry
When Willie Sutton, the famous bank robber, was asked why he robbed banks, he
immediately replied, "Because that's where the money is." The same is true of the
interest and attention companies are paying to cross-selling and upselling. There is
indeed money in these practices, and lots of money if and when they're performed
correctly. The result of properly implementing cross-selling and upselling techniques
is not only more sales today, but it should also include greater customer satisfaction,
greater customer loyalty, and greater overall customer spending with your company.
Why is this the case? Effective cross-selling and upselling must be regarded by the
customer as appropriate in the context of the conversation they are having, or the
practices will be viewed as inappropriate, annoying and self-interested.
The implication of adding cross-selling and upselling to a company's inbound
operation is that some form of "selling" is already taking place. Let's first examine this
assumption so we can proceed with the same understanding.
In an inbound sales department, whether it is business-to-business or business-to-
consumer, the customer calls due to some catalyst. This may be the arrival of a
catalog, seeing an infomercial or commercial on 1V, hearing a radio announcement,
maybe even buying from this company in the past. In other words, each customer
calls because of a specific precipitating event. Callers may be interested in a specific
product or service, or they may have been encouraged to call for more general
information. In deciding to make the call, each customer has, to some degree,
identified himself or herself as having understood your offer and made the decision to
interact with a representative of your company. This call was no accident; there was a
lot of "intention" on the part of the caller. What happens next during the call may not
make as much sense.
The majority of companies I encounter as a consumer and as a consultant are
reactive. That is, the customer/ caller controls the dialog, and the TSR (telephone
sales or service representative) is reacting to the questions and cues of the caller. I, as
the customer, ask some questions about price, delivery and availability, and they are
answered by the TSR. If I say nothing, no one does. I then decide to either place an
order or not to, and get off the phone. Did any "selling" take place? Were my needs
understood? Was my future value to the company realized, and was it acted upon?
Was I entered into a database for future contact, regardless of whether or not I
bought during this call? Did the TSR offer me any information about the company,
the quality of the products, the future satisfaction I could expect in doing business
with them, the warranties, my value to them? Was the TSR enthusiastic about having
me join his or her family of happy and lifelong customers who received value far in
excess of the price of a product, or was there just a mechanical and monotone voice
doing his or her job?
When your customers know how much you care, they won't care how much you
know.
Selling is the communication of appropriate benefits to a customer with a specific set
of needs. To sell, you must uncover and understand these needs, and then relate how
these needs will be satisfied through the use of your products or services. When needs
are not uncovered or discovered, no real selling can take place. Instead, customers are
offered benefits and then they decide -- on their own -- whether or not their needs are
served.
Cross-selling and upselling should be easy to integrate into a "selling" organization,
while nearly impossible to successfully integrate into an "order-taking" organization.
When customers' needs are uncovered and understood, it is easy to recommend
additional products or enhanced products because you, the salesperson, know and
understand how and why customers would benefit from their use. When basic needs
are not known and understood, cross-selling and upselling become as much an order-
taking operation as the initial sale. It is again left up to customers to decide to make
additional purchases, and they must usually do so in both an informational and
emotional vacuum.
Why is there such difficulty in establishing a sales organization, rather than an order-
taking organization in the first place? I have several theories about this, and they
originate in my own behaviors and observations. My first observation is that in
America today, "sell" is a four-letter word. If you ask a group of people (as I
frequently do in my seminars) to use some words to describe politicians, some of the
answers that emerge are: crooks; swindlers; liars; thieves; dishonest; self-serving; and
many worse descriptives that I need not share with you in print. When I ask these
same groups to apply words to salesperson, the very same words appear. This is
unfortunate. I consider myself to be a salesman, and sales to be an honorable
profession. I believe the majority of salespeople to be honest and well-intentioned,
even if they are not properly trained or managed. Why are salespeople the brunt of
jokes, stereotypes and absurd characterizations in both literature and films? I don't
know the answer, but I do know that we as a society willingly accept these stereotypes
as legitimate for both salespeople and politicians.
As salespeople, we are sensitive to our stereotype, and perhaps we go overboard not
to reinforce it. Many salespeople are aware of their intrusive, high-pressure image, so
they become completely passive and dispassionate in their roles of interacting with
customers. Salespeople fear that if they try to help customers solve their problems by
suggesting they buy the product or service the customers called about in the first
place, the salespeople will be perceived as pushy or high-pressure. What a sin this is]
Can you imagine going to your doctor, and then having him or her be timid about
making a diagnosis or writing a prescription because he or she feared you would
question his or her motive or intention? That is the state of sales in America today.
Perhaps the place to start changing this stereotype is with salespeople themselves.
When working with companies to improve the results and performance of their TSRs,
I am quick to point out that performance is the result of three dynamics. These are:
1. Skills,
2. Attitudes,
3. Beliefs.
Skills are the specific tasks and actions you instruct your TSRs to perform. Skills are
usually transferred during some activity that takes the form of training. Attitudes are
how you'd like these activities to be done, and attitude is usually managed through
motivation and incentives.
If you keep on doing what you're doing, you're going to keep on getting what you've
gotten. For most of us, that just isn't good enough.
Most selling programs, upselling programs and cross-selling programs fall short in
altering TSRs' beliefs. If TSRs believe in the worst stereotype of themselves as
salespeople, how can they possibly be asked to perform tasks they themselves
despise? If they believe themselves to be intrusive rather than caring, or high-
pressure rather than enthusiastic to solving customers' problems, or self-interested or
greedy rather than interested in recommending the best solution, how can they do the
job you ask of them?
The task for managers is to first understand your own beliefs about sales and the
actions you want people to take. You then need to understand how TSRs will regard
these actions in context with their beliefs about themselves and how they perceive
themselves treating other human beings. Only after you alter TSRs' beliefs into a
context of consultative selling, problem solving, service, and valuing customers as
"customers-for-life" can you ask them to do what they regard as unthinkable today--
sell the customers and solve their problems. People will rarely try to do something at
which they believe they will fail. We, as human beings, want to be consistent in our
beliefs and actions.
What the mind can conceive, man will achieve]
There was a time it was believed that it was impossible for a human being to break the
four-minute mile. Then one day Roger Banister did just that. What was amazing is
not that Roger Banister did it, but rather how many other runners did it immediately
after word got out that he had done it. Was it a change in human physiology? Hardly.
It was a change in beliefs. For us to be successful, we must not only believe that
something is possible, but we must also believe that we are capable of doing it. Too
often we focus on the exception. We are all too aware of irate and irrational
customers, and while their feelings need to be understood and corrected, the large
majority of our customers aren't like them. Let's focus our attention on true
consultative selling, on communicating with and understanding our customers, and
equally important, on giving them the chance to understand us. Let's make sure they
understand our commitment to lifetime customers -- not only is that good business
from an economic sense, it feels good. Let's make sure they understand our
commitment to customer satisfaction. Again, we know that it is cheaper to keep a
customer than gain a new one, and it feels better. Let's begin by selling customer
satisfaction. Let's understand our customers and their reasons for calling us in the
first place, and then cross-selling and upselling can be smoothly integrated into the
program.
Larry Levine is president and owner of LML Consulting Group. He is a sales trainer,
business consultant and motivational speaker and has managed sales forces for
multinational corporations, as well as consulted with dozens of companies to assist
their business growth.
Copyright Technology Marketing Corporation Dec 1996. Provided by ProQuest LLC. All inquiries
regarding rights or concerns about this content should be directed to Customer Serv~ce. For
permission to reuse this article, contact Cof:!yJjght Clearance Center.
HighBeam Research is operated by Cengage Learning.© Copyright 2018. All rights reserved.
www.highbeam.com