Opinion
No. 1:16 CV 9788
10-29-2019
MAUI JIM, INC., Plaintiff and Counterclaim Defendant, v. SMARTBUY GURU ENTERPRISES, MOTION GLOBAL LTD., SMARTBUYGLASSES SOCIETÁ A RESPONSABILITÁ LIMITATA, SMARTBUYGLASSES OPTICAL LIMITED, Defendants and Counterclaimants.
MEMORANDUM OPINION AND ORDER
MARVIN E. ASPEN, District Judge:
Presently before us is Defendants and Counterclaimants SmartBuy Guru Enterprises, Motion Global Ltd., SmartBuyGlasses Societá a Responsabilitá Limitata, and SmartBuyGlasses Optical Limited's (collectively "SmartBuyGlasses" or "SBG") motion to exclude survey and testimony of Brian M. Sowers (Dkt. No. 417.) For the following reasons, we deny Defendants' motion to exclude.
BACKGROUND
Plaintiff Maui Jim, Inc. ("Maui Jim") retained Brian Sowers of Applied Marketing Science, Inc. to run a survey to test Maui Jim's allegation that SBG's website gives consumers the impression SBG is an authorized Maui Jim retailer. (Pl. Mem. in Opposition to Def. Mot. to Excl. ("Pl. Mem.") (Dkt. No. 441) at 1.) Defendants argue Sowers' methodology in conducting this survey was "so flawed that this Court should exclude it and Mr. Sower[s'] related testimony." (Def. Mem. in Support of Mot. to Excl. ("Def. Mot.") (Dkt. No. 418) at 1.) SBG relies heavily on the expert report and deposition testimony of Dr. Howard Greenwald to criticize Sowers' survey. (Id.) Dr. Greenwald has a PhD in Sociology and has published extensively in the field of public health, including research and survey design in public health. (Pl. Mem. Ex. A (Dkt. No. 441-1) at 1-2.)
SBG puts forward three arguments against Sowers' survey: (1) the term "authorized retailer" is too ambiguous to give reliable survey results; (2) the "net confusion" survey control method invalidates the results; (3) the survey does not replicate realistic shopping conditions or use a proper sample. (Def. Mem. at 1-2.) Plaintiff responds that "authorized retailer" is a term of art that any ordinary person understands in the same way as Sowers used the term. (Pl. Mem. at 7-11.) Plaintiff defends the "net confusion" approach as the common practice in survey design for trademark cases. (Id. at 11-12.) Finally, Plaintiff explains the survey design replicated the shopping experience as much as possible and accurately reflected SBG's actual consumer pool; they also point out Dr. Greenwald had not seen the survey design himself before preparing his critique. (Id. at 13-15.)
SBG argues the purported errors in Sowers' survey design mean we should exclude the survey and related testimony under either Rules 403 or 702. Fed. R. Evid. 403 and 702, (Def. Mem. at 1.) Defendants argue the Court should exclude unreliable expert testimony in its "gatekeeping function" under Rule 702 and/or the Court should consider the unreliable survey evidence as "likely to confuse a jury" and therefore prejudicial under Rule 403. (Id.)
LEGAL STANDARD
"It is the job of the district court to ensure that the expert's opinion is reliable and relevant to the case, and thus, the district court is given broad discretion to do so." U.S. v. Young, 316 F.3d 649, 656 (7th Cir. 2002). "An expert must offer good reason to think that his approach produces an accurate estimate using professional methods, and this estimate must be testable." Durkin v. Equifax Check Servs., Inc., 406 F.3d 410, 421 (7th Cir. 2005) (quoting Zenith Elecs. Corp. v. WH-TV Broad. Corp., 395 F.3d 416, 419 (7th Cir.2005)).
"At the summary judgment stage, 'the judge's function is not ... to weigh the evidence and determine the truth of the matter but to determine whether there is a genuine issue for trial.'" Illinois Temale Co. v. El-Greg, Inc., No. 16 C 5387, 2018 WL 1534971, at * 7 (N.D. Ill. Mar. 29, 2018) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S. Ct. 2505, 2510 (1986)). The trial court's ultimate conclusion on the likelihood of confusion in a trademark case is a finding of fact. AHP Subsidiary Holding Co. v. Stuart Hale Co., 1 F.3d 611, 616 (7th Cir. 1993). Conclusion of that question of fact must be "approached with great caution." AutoZone, Inc. v. Strick, 543 F.3d 923, 929 (7th Cir. 2008) (quoting AHP Subsidiary Holding, 1 F.3d at 616).
ANALYSIS
We find no merit in Defendant's claim that Sowers' survey results were so unreliable as to warrant exclusion without a Daubert hearing. Evidence is relevant if it "has any tendency to make a fact more or less probable than it would be without the evidence" and "the fact is of consequence in determining the action." Fed. R. Evid. 401(a) and (b). Expert witnesses' specialized knowledge is admissible if "it will help the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702(a). We consider several factors, including "the expert witness's experience in a particular field," "[his] helpfulness to the jury," and "reflect reliable application" of "reliable principles and methods." Young, 316 F.3d at 656. Survey evidence is admissible and reliable when it does not contain leading or suggesting questions. Muha v. Encore Receivable Management, Inc., 558 F.3d 623, 625-26 (7th Cir. 2009). A properly designed control group is also vital in a survey intended to draw inferences about consumer behavior. See DeKoven v. Plaza Associates, 599 F.3d 578, 580 (7th Cir. 2010) (holding the same in the context of a debt collection letter survey).
One fact at issue in trademark infringement claims is whether consumers in the relevant market confuse the alleged infringer's mark with the complainant's mark. AHP Subsidiary Holding, 1 F.3d at 615 (citing Forum Corp. of North Am. v. Forum Ltd., 903 F.2d 434, 439 (7th Cir. 1990)). The Seventh Circuit has repeatedly allowed survey evidence using only photographic comparison to determine consumer confusion. Id. at 616; Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947, 959-60 (7th Cir. 1992) (endorsing reliability of this sort of survey evidence in a dispute about Gatorade); McGraw-Edison Co. v. Walt Disney Productions, 787 F.2d 1163, 1172 (7th Cir. 1986). Opinions compiled in an appropriate consumer survey are one form of expert testimony that can help determine confusion. See Taylor v. Cavalry Inv., 365 F.3d 572, 575 (7th Cir. 2004). A properly conducted consumer survey is one method of proving likelihood of mark confusion. See Eli Lilly & Co. v. Natural Answers, Inc., 233 F.3d 456, 464-65 (7th Cir. 2000) (pointing to a statistically reliable consumer survey as an appropriate method for proving mark confusion). One way to improve the reliability of consumer surveys is to include a "don't know/unsure" option. DeKoven, 599 F.3d at 581.
We will discuss each of Defendants' arguments against the reliability of Sowers' survey in turn. Because we ultimately do not find them persuasive enough to render Sowers' survey unreliable, we deny Defendants' motion to exclude.
I. The Term "Authorized Retailer" is Not So Ambiguous as to Taint Sowers' Results
Defendants argue the phrase "authorized retailer" as used in Sowers' survey is ambiguous and subject to two interpretations, only one of which points towards trademark confusion. (Def. Mem. at 3-4.) SBG attempts to distinguish between retailers Maui Jim authorized to sell its products from retailers permitted by law to sell its product. (Id. at 6.) Plaintiffs argue this is a distinction without a difference in the context of trademark law, and that Dr. Greenwald could not even distinguish the terms. (Id. at 7-8.) They also argue did define "authorized retailer" as "a retailer who is legally authorized to sell Maui Jim goods online" in the relevant survey question. (Id.; Pl. Mem. Ex. B ("Greenwald Dep.") (Dkt. No. 441-2) at 138:3-9.)
The phrase "authorized retailer" is not ambiguous in the context of this survey. Any daylight between permission to sell lawfully and Maui Jim's permission to sell is tough to parse here; not even Dr. Greenwald seems capable of concretely describing the distinction. (Greenwald Dep. at 140:23-141:3.) For example, Dr. Greenwald states that words only become ambiguous when they pass a certain threshold of uncertainty, while admitting that he is unsure whether the phrase "authorized retailer" passes this threshold. (Id. at 144:2-145:15.) There is nothing to show the failure to conduct a pretest on the phrase "authorized retailer" actually would taint the survey results. Instead, Dr. Greenwald asserts "[he] knows what leads to invalidity" and "had a number of different clients in the real world," so therefore he is capable of determining which words must be subject to pre-screening. (Id. at 144:13-21.) In fact, Dr. Greenwald himself conducted none of the analyses he criticized Sowers for excluding. Dr. Greenwald's bare appeal to experience does not substitute for empiricism, especially given Dr. Greenwald's apparent emphasis on public health surveys and lack of previous experience in consumer confusion surveys. (Id. at 68:24-72:10.) He simply asserts that the phrase "authorized retailers" is necessarily ambiguous because he "know[s] what can lead to invalidity" and "authorized retailer is technical language . . You just know that the survey respondent isn't going to know much about it." (id. at 144:15-145:15.) In other words, Dr. Greenwald's report, standing alone, is not enough to convince us that Sowers' survey is so unreliable as to require exclusion under Rule 702. On the contrary, it appears similar to the types of surveys this circuit routinely embraces. See, e.g. Taylor, 365 F.3d at 575; Sands, 978 F.2d at 959-60; DeKoven, 599 F.3d at 581.
II. "Net Confusion" is Sufficiently Established as a Control Mechanism
A survey designed to estimate likelihood of confusion usually includes a proper control. THOIP v. Walt Disney Co., 690 F. Supp. 2d 218, 240 (S.D.N.Y. 2010) (citing Shari Seidman Diamond, Reference Guide on Survey Research, in Reference Manual On Scientific Evidence ("Diamond on Survey Research") at 229, 257 (Federal Judicial Center 2d ed.2000)); see also DeKoven, 599 F.3d at 580 (applying the same logic to debt collector confusion claim). A control is designed to estimate the degree of background error in the survey, which provides a proper benchmark for "determining whether a likelihood of confusion estimate is significant or merely reflects flaws in the survey methodology." THIOP, 690 F. Supp. 2d at 240 (citing Diamond, at 257). A control typically should share "as many characteristics with the experimental stimulus as possible, with the key exception of the characteristic whose influence is being assessed." Id. A survey control that involves changing multiple relevant components does not effectively serve this function. Id. at 240-41. Thus, consumer surveys should use controls designed to reduce error to the extent possible. Id. at 241; see also DeKoven, 599 F.3d at 580-81. Internal controls differ from external controls because internal controls use comparisons within the survey group to multiple tested and control questions, while external controls use multiple groups of people, only some of whom are exposed to relevant stimuli. Jacob Jacoby, Experimental Design and the Selection of Controls in Trademark and Deceptive Advertising Surveys, 92 Trademark Rep. 890, 895 (2002). "Common forms of internal controls used in trademark and deceptive advertising studies include other questions (such as those asked about meanings presumably not contained in an allegedly misleading ad) and other stimuli." Id. Nevertheless, "while the fact that a survey used a control that could have been 'stronger' or 'better' may mean it is entitled to less weight, it does not mean that the survey does not provide relevant information." Gucci Am., Inc. v. Guess?, Inc., 831 F. Supp. 2d 723, 740 (S.D.N.Y. 2011).
Defendants' objections to Sowers' study go to weight, not admissibility. Although it is true a study that contains no controls falls below standards of reliability, there is no hard-and-fast requirement that a certain proportion of variation be explicable before a study becomes admissible. See THIOP, 690 F. Supp. 2d at 241. Dr. Greenwald presents ability to explain variation between covariates as the hallmark of reduction in "noise," but he makes no baseline comparison to other studies ruled inadmissible in this circuit (or any court for that matter). Without such a baseline we cannot say whether Sowers' correlation coefficient deviates significantly from standard practice in the field of trademark confusion or consumer survey design. Further, Sowers' used net confusion based on a "within control," a common technique for controlling statistical noise in consumer surveys. Jacoby, 92 Trademark Rep. at 895. Thus, without a compelling reason justification for excluding Sowers' survey evidence entirely, we are inclined to consider Dr. Greenwald's critiques of the survey as going to weight of the evidence.
III. The Survey Sample and Marketplace Conditions were Sufficiently Proper
Surveys should replicate the marketplace conditions as well as possible in order to control potential noise. THIOP, 690 F. Supp. 2d at 240. No control is perfect, so reliability does not require exact simulation of the situation at issue in a trademark suit. Id. at 239 n. 148. Defendants argue the Maui Jim customer is usually regarded as seeking "functional elegance" and that Sowers did not properly construct a group of consumers using this frame. (Def. Mem. at 10-11.) Defendants' argument does not match the actual survey design, given the only measurable characteristics they describe as equating to functional elegance are age and disposable income. (Id. at 11.) Sowers' survey specifically defined potential purchasers using both age and willingness to expend income on sunglasses. (Pl. Mem. at 13-14.) Plaintiff correctly points out willingness to pay for sunglasses actually adheres more closely than the disposable income proxy that Defendants suggest, since even those with low income may prefer "functionally elegant" sunglasses. (Id. at 13.) Nothing about the construction of the "potential purchaser" group appears out-of-step with even Defendants' own characterization of the appropriate group definition. In other words, Sowers' survey defined the customer using the same characteristics Defendants' claim describe their customer, only with greater particularity than the Defendants' description. Defendants' argument is frankly an attempt to use marketing terminology to disguise a meritless assertion.
CONCLUSION
Defendants have made several arguments that go to the weight of Sowers' survey should receive, which are fair game for a jury. Nevertheless, we do not believe Sowers' survey is so unreliable as to exclude it. Thus, we deny Defendants' motion to exclude. (Dkt. No. 418.) It is so ordered.
/s/_________
Honorable Marvin E. Aspen
United States District Judge Dated: October 29, 2019
Chicago, Illinois