Lulu Enterprises, Inc. v. N-F Newsite, LLC et al

Filing 112

Memorandum in Support re #111 MOTION in Limine to Exclude Plaintiff's Surveys and Expert Testimony Under Daubert filed by N-F Newsite, LLC. (Attachments: #1 Appendix Unreported Cases) (Silver, Hayden)

Lulu Enterprises, Inc. v. N-F Newsite, LLC et al

Doc. 112

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 1 of 16

IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF NORTH CAROLINA WESTERN DIVISION Case No. 5:07-CV-00347-D LULU ENTERPRISES, INC., Plaintiff, vs. HULU, LLC, et al. Defendants. MEMORANDUM OF LAW IN SUPPORT OF DEFENDANT'S MOTION TO EXCLUDE PLAINTIFF'S SURVEYS AND EXPERT TESTIMONY

This Court's gate-keeping function requires the exclusion of the surveys and reports of Plaintiff's expert witnesses: (1) Hal Poret and Robert Reitter (of Guideline Research); and (2) Kent Van Liere (of NERA Economic Consulting) (collectively "Plaintiff's Expert Reports"). In addition to being untimely,1 Plaintiff's Export Reports are unreliable and non-probative on the issue of consumer confusion, failing to satisfy the requirements under the Supreme Court's Daubert case, as well as the requirements set forth in Rule 702 of the Federal Rules of Evidence. I. PLAINTIFF'S EXPERT REPORTS FAIL TO SATISFY THE DAUBERT STANDARD FOR ADMISSIBLE SURVEY EVIDENCE Plaintiff cannot satisfy its burden of proving, by a preponderance of the evidence, that Plaintiff's Expert Reports are admissible or reliable. Recognizing the heightened power of "expert" evidence to mislead, the United States Supreme Court held in Daubert v. Merrell Dow Pharmaceuticals, Inc., that a trial judge has a "gatekeeping obligation" to place appropriate

Defendant has opposed Plaintiff's motions for leave to file these declarations and reports, and on October 12 filed a motion in limine, on the basis of the reports being untimely. See Def.'s Mot. in Limine to Exclude Pl.'s Survey and Focus Group Declarations and Reports; and Def.'s Mem. of Law in Opp. to Pl.'s Mot. for Leave to File Decl. of Hal L. Poret and In Supp. of Hulu's Mot. in Limine [Dkt. 107 and 108].

1

US2000 10376177.5

Dockets.Justia.com

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 2 of 16

limits on the admittance of scientific evidence. 509 U.S. 579, 579-80 (1993); see also Cooper v. Smith & Nephew, Inc., 259 F.3d 194, 199 (4th Cir. 2001) ("Under Federal Rule of Evidence 702, trial judges act as gatekeepers to `ensure that any and all scientific testimony. . . is not only relevant, but reliable.'") (quoting Daubert, 509 U.S. at 588).2 The "proponent of a consumer survey has the burden of establishing that it was conducted in accordance with accepted principles of survey research" and that the Daubert requirements are satisfied. See Nat'l Football League Props. v. New Jersey Giants, 637 F. Supp. 507, 513 (D.N.J. 1986); see also Daubert, 509 U.S. at 592 n.10. The court should ensure that an expert "employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." See Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999) (holding that Daubert applies to non-scientific experts as well); see also 3 J. Thomas McCarthy, McCarthy On Trademarks & Unfair Competition § 11:88 ("McCarthy"). Specifically, for a survey to be admissible, it must have been conducted in accordance with accepted principles of survey research, which have been amplified for the Federal Judicial Center by Dr. Shari Diamond in Reference Guide on Survey Research, Reference Manual on Scientific Evidence (2d ed. 2000) ("Diamond, Reference Guide").3 A. The Unreliable and Non-Probative Guideline Survey

Guideline screened participants for internet access (but did not require high-speed access), and allowed participation for an array of people who watch video clips (including only homemade, non-professional video, such as available on YouTube.com). Then, Guideline:
Daubert sets out four nonexclusive factors for admissibility of scientific expert testimony: whether the theory or technique "(1) can be or has been tested; (2) has been subjected to peer review and publication; (3) has a high known or potential rate of error and whether there are standards controlling its operation; and (4) technique enjoys general acceptance within a relevant scientific community." 509 U.S. at 592-94. A copy of the Reference Guide on Survey Research is available on the Federal Judicial Center's website at: http://www.fjc.gov/library/fjc_catalog.nsf.
3

2

2

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 3 of 16

· · ·

showed the home pages from the following three websites: Lulu.com, Createspace.com, and Veoh.com (from which the respondent could navigate); showed a made-up Yahoo! advertisement for Hulu.com (that Guideline supplied), and asked, inter alia, "Do you think this advertisement is for one of the websites that I showed you earlier, or do you think not, or don't you know?" (Guideline, Question 1.a).

Guideline's poorly conceived survey fails to satisfy the requirements of Daubert, supra, and should be excluded because the survey: · Uses an improper universe by including participants who were o not likely to be Hulu's customers o without high-speed internet access Fails to replicate the marketplace conditions by using o an artificial testing environment o an improper stimulus Uses an inappropriate format o allowing sequential, essentially side-by-side comparison o creating an inappropriate exposure to Lulu.com, which has little awareness o creating an inappropriate demand effect Fails to comply with established, industry guidelines for market research (due to absence of tables, verbatims, etc.), which are essential to evaluating reliability.

·

·

·

For each of these reasons, as well as the cumulative impact of these methodological flaws, the Guidelines report should be excluded; it simply does not measure a likelihood of confusion in any meaningful way.4 For the proposition that the Guideline (Poret/Reiter) study must be rejected under Daubert principles, this Court need only read and apply four cases and Diamond's Reference Guide.

For example, the fact that the level of "mis-measurement" is so high for the VEOH site confirms that respondents answers are not trademark confusion (based on names), but rather simply forced association with limited choices. See Supplemental Decl. of Dr. Gerald L. Ford ¶¶ 10-11 (describing use of "invalid research design," and apparent failure to measure trademark confusion); see also id. ¶ 4 ("Guideline offers a survey design that has been repeatedly rejected by courts").

4

3

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 4 of 16

1.

Improper Survey Universe: Citizens Fin. Group, Inc. v. Citizens Nat'l Bank of Evans City, 2003 WL 24010950 (W.D. Pa. Apr. 23, 2003), aff'd, 383 F.3d 110, 118-21 (3d Cir. 2004).

It is axiomatic that a defendant's prospective customer base is the proper universe for a confusion study. Zimmerman v. Nat'l Ass'n of Realtors, 2004 T.T.A.B. LEXIS 180 (T.T.A.B. 2004).5 Here, however, the universe includes anyone who in the past or the next three months is likely to use the Internet to "watch or download video clips." This is the expansive YouTube universe, not the more circumscribed Hulu universe. In other words Guideline did not limit the survey participants to Hulu's customers, i.e., people who would watch premium television shows or feature films, but instead permitted respondents who had an expressed interest in watching any type of video clips (i.e, homemade, YouTube type videos), or downloading any movies (i.e., even if not a feature film).6 In addition, the "practical reality" is that consumers will need highspeed internet access to successfully view hulu.com (Decl. of Jason Kilar ¶ 25), but Guideline allowed participation based on any internet access, including a dial-up connection.7

Whether the survey "population was properly. . .defined" and whether "the sample chosen was representative of that population" is critical. The Manual for Complex Litigation, Fourth § 21.493 (2004). "A survey that provides information about a wholly irrelevant universe of respondents is itself irrelevant." See Diamond, Reference Guide, at 241 ("[E]ven if the proper questions are asked in the proper manner, if the wrong persons are asked, the results are likely to be irrelevant."); see also McCarthy § 32:159. More commonly, the survey universe is either "underinclusive or overinclusive." See Diamond, Reference Guide, at 241. A quick review of questionnaires (on Sunday, October 14) indicated that over 20% of the survey universe ­ 85 of 441 ­ qualified without indicating that they had or would watch TV shows. In fact, over 10%­ 47 of 441 ­ qualified even though they would not watch TV shows or films on an internet site. Demonstrating the importance of flaws in the universe, a significant number of the respondents who are labeled as "confused" fall into this category of non-TV/film viewers (including at least 1023, 1105, 2002, 2003, 2016, 2047, 2080, and 2086). As reported by The Pew Internet & American Life Project, an affiliate of The Pew Charitable Trusts, 47% of adult Americans have broadband connection at home as of early 2007. See http://www.pewinternet.org/pdfs/PIP_Broadband%202007.pdf (the "Pew Report"). Among individuals who use the internet at home, only 70% have broadband connection, while 23% use dialup. Id. Dialup users are generally not expected to be watching high-quality video content like premium TV shows, and also are less "savvy and sophisticated internet users." See Kilar Decl. ¶ 26; see also Pew Report 10 ("In addition to using the internet more frequently than individuals with dialup access, broadband users also
7 6

5

4

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 5 of 16

The "failure to conduct the survey within [defendant's] universe" mandates that the Guideline survey here be rejected for the same reason that a study by Reitter was rejected in Citizens Financial Group, Inc. v. Citizens National Bank of Evans City, 2003 WL 24010950, at *3-*5 (W.D. Pa. Apr. 23, 2003), aff'd on these grounds, 383 F.3d 110, 118-21 (3d Cir. 2004). In that case, the Court held that "the problems with the Guideline Research Survey are so fundamental and basic that the survey is stripped of any significant probative value. Therefore, I am convinced that the Guideline Research survey is not simply ill-tailored (so as to go to the weight), but is fatally flawed due to an irrelevant and improper universe, such that it must be excluded from evidence at trial." Id. at *5 (emphasis added).8 This is not an arcane or arbitrary principle. Rather, as stated in Weight Watchers Int'l, Inc. v. Stouffer Corp., 744 F. Supp. 1259, 1272-73 (S.D.N.Y. 1990), where the plaintiff did not limit the universe to defendant's "diet frozen entrée" market, "[r]espondents who are not potential consumers may well be less likely to be aware of and to make relevant distinctions when reading ads than those who are relevant consumers." Accordingly, it is generally appreciated, as observed in Zimmerman, that divergent universes may produce diametrically different results, and that "[a] survey that provides information about a wholly irrelevant universe of respondents is itself irrelevant."9

participate in a wider range of online activities. It is perhaps not surprising that broadband users exhibit greater rates of participation in bandwidth intensive activities . . . that are cumbersome and time consuming at dialup speeds."). Other surveys by Dr. Reiter that have been heavily criticized, but are not discussed below include: (1) Classic Foods Int'l Corp. v. Kettle Foods, Inc., 468 F. Supp. 2d 1181, 1193 (C.D. Cal. 2007). (The "[Reitter genericism] survey . . . has several flaws in its methodology. It uses a circular definition for what a brand name is, it fails to screen subjects to see if they understood what a brand is, and it fails to identify or account for any order bias that might have been produced."); and (2) Pilot Corp. of Am. v. Fisher-Price, Inc., 344 F. Supp. 2d 349, 359 (D. Conn. Nov. 9, 2004) ("the Reitter Survey suffered from a small sample size").
8

5

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 6 of 16

2. Inappropriate Survey Format: National Distillers Products Co., LLC v. Refreshment Brands, Inc., 198 F. Supp. 2d 474, 482-84 (S.D.N.Y. 2002). Guideline respondents at the mall were shown Lulu.com, Createspace.com and Veoh.com, and then immediately thereafter, shown a made-up Yahoo! advertisement for Hulu.com, and asked: "Do you think this advertisement is for one of the websites that I showed you earlier, or do you think not, or don't you know?" Such a format "provide[s] the respondent with an explicit set of responses [and] may remind respondents of options that they would not otherwise consider or which simply do not come to mind as easily."10 Not surprisingly, the format produces high levels of guessing and yea-saying and has attracted adverse reviews from commentators and courts.11 As captured in the verbatim for respondent 2087, "If I had to pick one, I'd say its hulu.com." As Dr. Ford observes:

Diamond, "Reference Guide," at 241. See J&J Snack Foods, Corp. v. The Earthgrains Co., 220 F. Supp. 2d 358, 370-72 (D.N.J. 2002) ("It is undisputed that [plaintiff] sells all of its frozen cookie dough product with the `BREAK & BAKE' mark through its food service distributors to distributors who offer the product to fund raising organizations. . . . [T]he universe [chosen for the study, however,] consisted of adults shopping at one of four malls in the United States who either `do at least some of the grocery shopping for their household,'. . .`have purchased refrigerated cookie dough in the past and/or intend to purchase refrigerated cookie dough in the future. . . .' This universe renders the study unreliable. . . .") Just as a survey universe of respondents who "regularly consume `liquor, such as gin, rum, tequila, or vodka'" is not representative of purchasers of vodka coolers, see Nat'l Distillers Prods. Co., LLC v. Refreshment Brands, Inc., 198 F. Supp. 2d 474, 484 (S.D.N.Y. 2002), a universe of individuals who are not going to be watching premium television shows and do not have high-speed internet access, is not representative of Hulu's potential consumers that must have access to high-speed internet service. Diamond, "Reference Guide" at 251. At minimum, the format "often indicate[s] to respondents areas of interest to the surveyor." Richard J. Leighton, "Using Daubert-Kumho Gatekeeping to Admit and Exclude Surveys in Lanham Act Advertising and Trademark Cases," 92 Trademark Rep. 743, 781 (2002). See R. Bradlee Boal, "Techniques for Ascertaining Likelihood of Confusion and the Meaning of Advertising Communications," 73 Trademark Rep. 405, 422 (1983), noting that the Squirt question, "Do you think these two products are made by the same or different producers?," is not neutral, but "strongly suggests a possibility that might not have occurred to the interviewees ­ that the products are made by the same company." "[T]he mere putting of [the] question creates the impression of a relationship." Kargo Global, Inc. v. Advance Magazine Publishers, Inc., 2007 U.S. Dist. LEXIS 57320, *26 (S.D.N.Y. Aug. 6, 2007).
11 10

9

6

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 7 of 16

Surveys which employ leading or suggestive questions do not provide an appropriate measurement for likelihood of confusion in Lanham Act matters. In the instant matter, the Guideline survey was clearly suggestive. [The Guideline] procedure not only suggested that survey respondents should find a connection between Hulu and one of the websites shown earlier but specifically with the website with the closest similarity to one they had seen earlier. Rather than measuring any actual likelihood of confusion the Guideline survey questions and procedures generated "demand effects" by suggesting to the survey respondents the existence of a connection between the internet sites and the advertisement that the survey respondents would not have made on their own. Supplemental Ford Decl. ¶¶ 7-8. The format was rejected, for example, in National Distillers Products Co., LLC v. Refreshment Brands, Inc., where the Court noted that absent its display, "respondents would have been unfamiliar with [the allegedly infringed product] due to [its] very limited distribution network and weak sales." 198 F. Supp. 2d 474, 482-84 (S.D.N.Y. 2002). Defendant's previously submitted recognition survey in this case establishes that 98% of defendant's customers do not have even aided awareness of plaintiff. See Ford Decl. ¶ 3 ("no significant degree of awareness"). Plaintiff's improperly leading survey format must thus be here likewise rejected. M.D. On-Line v. WebMD Corp., 2005 U.S. Dist. LEXIS 23563, at *21-23 (D.N.J. Oct. 6, 2005). 3. Failure to Replicate the Marketplace: Simon Property Group L.P. v. MySimon, Inc., 104 F. Supp. 2d 1033 (S.D. Ind. 2000). "The huge volume of information available on the Internet can be both a source of confusion and a means for avoiding that confusion. Consumer surveys dealing with confusion on the Internet must keep in mind these important features of the Internet experience." Simon Prop. Group L.P. v. MySimon, Inc., 104 F. Supp. 2d 1033, 1038 (S.D. Ind. 2000). Particularly, a survey as to the Internet experience "must attempt to replicate the thought processes of consumers encountering the disputed mark or marks as they would in the marketplace." Id. at

US2000 10376177.5

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 8 of 16

1038. Accordingly, a "survey [that] does not ask respondent to view actual search engine results containing hyperlinks to . . . sites in a way that might actually occur in the marketplace [is] nothing more than a meaningless memory game or word association exercise that bears no relationship to the marketplace." Id. at 1043 (emphasis added). Here, Guideline simply showed each respondent the Lulu.com website, and then two additional ones chosen by Guideline: Createspace.com and Veoh.com. Within seconds, of seeing Lulu.com, Createspace.com and Veoh.com, each participant was also shown a made-up Yahoo! advertisement for Hulu.com. The likelihood that such a sequence would occur in the marketplace is patently nonexistent. See Supplemental Ford Decl. ¶ 9 ("The Guideline survey structure does not appear to be based upon any marketplace conditions."). In fact, the Guideline approach is inconsistent with the only evidence in the record about how consumers use and locate information on the internet. See Decl. of Erich Joachimstaler ¶27 (describing "leanforward experience that is highly involving"). The made-up advertisement for Hulu.com is inherently misleading and inaccurate.12 A significant number of respondents indicated confusion because of the ads' look and feel,

First, Hulu is not going to do any advertising, so its mark never would appear in that context. See Kilar Decl. ¶ 37. Hulu's "current intent is that Hulu will not promote itself by doing any advertising." Id. Second, when Hulu does appear on a distribution partner site (such as Yahoo!), it will be in the form of "a video player branded with the `Hulu' name and offering the same premium content available on Hulu's own site." Id. ¶ 64. As a result, Guideline's mock-ad contains advertising that Hulu does not currently intend to use, and Guideline fails to limit the use of the HULU mark to video player. Because the stimulus shown to interviewees did not replicate the marks in context as "presented" in the marketplace, the survey results are not relevant. See, e.g., Playtex Prods., Inc. v. Georgia-Pacific Corp., 390 F.3d 158, 168 (2d Cir. 2004). Third, the site Hulu.com will have a distinctive look that prominently features well-known television programming and feature films. By selecting a "Yahoo! ad," Guidelines chose to completely ignore that aspect of Hulu's business. In other words, none of this analysis applies to Hulu.com. Fourth, the stimulus that Guideline's selected to use obviously and artificially included lesser known television programming, which distorts the results. Rather than using each show identified on the Hulu.com website, they used either new or generally lesser-known than the shows and skipped over others (e.g., The Office, 24, Prison Break, Heroes, 30Rock, etc.). See, e.g., Nat'l Distillers Prods., 198 F. Supp. 2d at 484 (excluding survey because, inter alia, the cooler bottles used as a stimulus where empty when shown to respondents, but would have been full when sold in the marketplace).

12

8

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 9 of 16

format, or what was listed but not the name which reinforces and confirms the magnitude of an improper stimulus.13 4. Inappropriate Sequential Presentation: Kargo Global, Inc. v. Advance Magazine Publishers, Inc., 2007 U.S. Dist. LEXIS 57320 (S.D.N.Y. Aug. 6, 2007). Respondents viewed the websites and ad literally or effectively side-by-side. The

numbers of cases rejecting side-by-side displays, when the objects of a study do not so appear in the marketplace, are legion.14 The Guideline study must thus be rejected for the reasons

articulated in Kargo Global, Inc. v. Advance Magazine Publishers, Inc., 2007 U.S. Dist. LEXIS 57320, at *26 (S.D.N.Y. Aug. 6, 2007): Here, the back-to-back, or seriatim, display of the Cargo and Kargo marks did not approximate conditions that consumers would encounter in the marketplace. The products at issue (Kargo's wireless services and Advance's men's shopping magazine, respectively) were not competing. Although some overlap existed in the demographic makeup of both Kargo's and Cargo's target audiences, the companies were engaged in different businesses. . . . Kargo has offered no data or other evidence to support the proposition that prospective customers were likely to encounter Kargo's trademark a short time after seeing Cargo magazine. Here, therefore, as in Beneficial Corp. v. Beneficial Capital Corp., 529 F. Supp. 445, 450 (S.D.N.Y. 1982), presenting two marks side-by-side or seriatim and asking if there is a "business connection," creates demand effects and elicits influenced responses that appear on their surface to favor a plaintiff. In fact, such a survey "establishes no more than that the names are similar, a factor as to which there can be little genuine dispute in any event, and that portions of the general public will make the reasonable assumption, that, in the absence of any other information, two companies with similar names are likely to have a business connection. However, this
See, e.g., Questionnaire nos. 1040 ("it looks like the set up"); 1053 ("recognize some of the graphics"); 2002 ("because it was advertising"); 2024 ("advertisement"); 2048 ("the pictures look similar"); 2064 ("because the ads"); 2086 ("because it's laid out almost the same"). See, e.g., Jordache Enters. v. Hogg Wyld, Ltd., 828 F.2d 1482, 1488 (10th Cir. 1987); Leelanau Wine Cellars, Ltd. v. Black & Red, Inc., 452 F. Supp. 2d 772, 783-84 (W.D. Mich. 2006).
14 13

9

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 10 of 16

proposition provides no indication of public reaction under actual market conditions, and we conclude that there is no meaningful evidence of actual confusion." Id. at 451 (emphasis added).15 Indeed, here, as in Riviana Foods Inc. v. Societe Des Produits Nestle S.A., 33 U.S.P.Q. 2d 1669, 1671 (S.D. Tex. 1994), "the failure of approximately 75% of those questioned to follow the leading suggestion [in the survey] and answer that the products were made by the same company is strong evidence that there is no confusion." 5. Inadequate Reporting: Diamond's Reference Guide on Survey Research

"A survey report generally should provide in detail . . . (6) a description of any special scoring (e.g., grouping of verbatim responses in broader categories); . . . (8) statistical tables clearly labeled and identified as to source of data, including the number of raw cases forming the base for each table, row, or column . . . ." Diamond, "Reference Guide" at p. 271 (noting that criteria were adapted from the Council on Am. Survey Res. Orgs., Code of Standards and Ethics for Survey Research § III.B.4 (1997)). Failure to supply this information substantially impairs a court's ability to evaluate a survey. Id. at 271, n.158 (citing Prudential Ins. Co. of Am. Sales Practices Litig., 962 F. Supp. 450, 532 (D.N.J. 1997)). The Guideline report fails to include any statistical tables or raw data in the form of verbatim responses for the participants (but for a few selective, self-serving quotes). See

Supplemental Ford. Decl. ¶ 12. Both Hulu and this Court do not have any ability to examine

In Louis Vuitton Malletier v. Dooney & Bourke, Inc., 340 F Supp. 2d 415, 445-46 (S.D.N.Y. 2004), rev'd on other grounds, 454 F.3d 108 (2d Cir. 2006), the court questioned Dr. Reitter's survey and discounted it on the grounds that was, as here, "essentially a reading test." Of course, it is well settled that a likelihood of confusion survey must replicate actual marketplace conditions to be relevant and probative of confusion. See McCarthy § 32:163; J & J Snack Foods, Corp. v. Earthgrains Co., 220 F. Supp. 2d 358, 371 (D.N.J. 2002) (excluding survey under Rule 702 because it did not imitate market conditions); see also Coherent, Inc. v. Coherent Techs., Inc., 935 F.2d 1122, 1126 (10th Cir. 1991) (finding district court did not clearly err in rejecting survey evidence that failed to simulate market conditions).

15

10

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 11 of 16

Guideline's conclusions, and independently evaluate the survey results.16 Given the nature of prior criticism,17 and the methodological flaws outlined above, Guideline's continuing failure to comply with well-established standards for survey research should result in the exclusion of the report.18 B. The Unreliable and Non-Probative Van Liere Report.

The Van Liere Report is based exclusively on the results of qualitative "focus groups" held in three locations involving a sum total of 40 individuals.19 This type of qualitative, focus group evidence is routinely rejected by courts, which find that such research does not offer statistically reliable evidence on the issue of likelihood of consumer confusion. In Scotts Co. v. United Industries Corp., 315 F.3d 264 (4th Cir. 2002), the Fourth Circuit reversed the grant of a preliminary injunction and found that the trial court abused its discretion by giving focus group evidence weight in determining whether consumers were

And those errors do exist even on the face of the report. See Guidelines Report, App. E (survey respondent #1024 included in "Hulu Live TV (Test)" cell even though survey number does not match; survey respondent #3003 shows up in two places, etc.). In Louis Vuitton Malletier, 340 F Supp. 2d at 446, involving a Reitter/Guidelines survey, the court was troubled by "several miscoded responses . . . concluding that "he survey improperly inflates the number of responses illustrative of `noise,' which in turn erroneously deflates the confusion figures."). Other Guideline surveys heavily criticized and/or excluded include Pfizer, Inc. v. Miles, Inc., 868 F. Supp. 437, 447-48 (D. Conn. 1994): The court did not give great weight to the survey because the Guideline expert used an inappropriate stimulus and "may have misinterpreted the [ ] verbatim responses"; Warner Lambert Co. v. McCrory's Corp., 718 F. Supp. 389 (D.N.J. 1989) (denying plaintiff's motion for preliminary injunction, finding that Guideline's survey "does not provide a sufficient basis in and of itself for a finding that plaintiff is likely to be able to prove consumer confusion." 718 F. Supp. at 399; Prince Mfg., Inc. v. Bard Int'l Assoc., Inc., 1988 WL 142407, at *7-8 (D.N.J. Dec. 22, 1988) (affording the survey "scant weight" because, inter alia, the survey failed to focus on the marks as used on the products at issue, failed to survey the defendant's "niche market," and used leading questions that "informed respondents"). By Dr. Van Liere's own admission, the focus groups at issue involved "open-ended discussion" among the participants. See Van Liere Report 4. Dr. Van Liere's report concedes, "[g]iven the time constraints for this project, it was not possible to conduct both a series of focus groups and then field a full quantitative survey." Id. Dr. Van Liere's Report, by his own admission, is therefore "based on the qualitative focus groups alone and represents his "initial evaluation." Id. at 2.
19 18 17

16

11

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 12 of 16

misled by defendant's allegedly misleading advertising. Id. at 277. The court expressed extreme skepticism as to whether focus group evidence should ever be admitted, noting that "the very nature of a focus group seems, to some extent, to limit its ability to identify the message an advertisement conveys to an individual consumer." Id. at 277 (emphasis added).20 Scotts is hardly alone in recognizing the fundamentally flawed nature of focus group evidence when determining the issue of consumer confusion.21 The inherent flaws of qualitative focus group research are compounded in the Van Liere research by many of the same defects that plague the Guideline research, including: · A Flawed Universe: The Van Liere incorrectly assumes that Hulu will allow video uploads (see Van Liere Report 5, citing a standard "terms and conditions" page), and erroneously uses that as a basis for qualifying the participants even though Hulu will not allow such content (see Kilar Decl. ¶¶ 21-23). It also is not limited to people who have high-speed internet access (see Van Liere Report 7);

The court in Scotts ultimately did not decide the "broader" question of whether focus group evidence is generally admissible because it found that the focus group evidence presented to it was unreliable and could not be considered probative on the issue of consumer confusion. Scotts, 315 F.3d at 277. Here, too, the particular focus group evidence is so flawed as to be inadmissible regardless of the general admissibility of focus group evidence. See, e.g., Merisant Co. v. McNeil Nutritionals, LLC, 242 F.R.D. 315 (E.D. Pa. 2007) ("the results of a focus group are inherently different than the results of a survey. [W]hereas a survey is designed to collect data from an appropriate numerical sample of individual respondents, focus groups are designed to gather data from `groups' of people assembled together for that very purpose."); Smithkline Beckman Corp. v. Proctor & Gamble Co., 591 F. Supp. 1229, 1239 (N.D.N.Y. 1984), aff'd 755 F.2d 914 (2d Cir. 1985) (noting "serious reservations about relying on [focus group evidence] as proof of confusion as to the source of the products on the part of an appreciable number of ordinarily prudent purchasers. [E]ven if the Court concluded that the focus group somehow recreated the atmosphere of the marketplace, which it clearly does not . . . any results obtained would be particular to that focus group and could not be used to predict the behavior of consumers as a whole."); Procter & Gamble Co. v. Colgate-Palmolive Co., 1998 WL 788802, at *80 (S.D.N.Y. Nov. 9, 1998), aff'd, 199 F.3d 74 (2d Cir. 1999) ("The focus group studies . . . do not meet the evidentiary standards for market studies, but even if admitted, purported statements made during the focus groups would be afforded limited weight because the survey sample was very small and neither the actual declarants nor the survey takers documenting such alleged statements testified at trial.").
21

20

12

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 13 of 16

·

A Flawed Stimulus: The sheets with the multiple names do not show the marks as used in the marketplace, but improperly use a name alone, which is a basis for complete exclusion. See Juicy Couture, Inc. v. L'Oreal USA, Inc., 2006 WL 1012939, at *24-26 (S.D.N.Y. Apr. 19, 2006) (criticizing use of 81/2 x 11 card with name, as opposed to actual product, as "fundamental flaws" failing to "replicate the market conditions"); see also Supplemental Ford Decl. ¶ 14 ("The manner in which the focus group participants were presented with the website addresses, in plain block letters on a sheet of paper, does not represent in any way the manner in which these marks would be encountered in the marketplace. As such, the data from the focus group regarding the relationship between the website addresses does not provide a valid measure of likelihood of confusion.");

·

Improper "Side by side" Appearance: The use of the names on a sheet results in what essentially is a matching test, rather than a test of likelihood of confusion (see, supra, at 7-9, discussing Guideline flaws); see also Supplemental Ford Decl. ¶ 15 ("inappropriate unless the marks are encountered side-by-side in the marketplace");

·

Leading and Suggestive Questions: The participants were asked to do the two exercises involving drawing a line, but in a manner that encouraged guessing without any control.22

"Surveys that record consumer impressions have a limited ability to answer questions about the origins of those impressions." Diamond, "Reference Guide" at p. 256. The consumer's response to any question may be the result of information or misinformation from sources other than the trademark the respondent is being shown. Id. "The more routine the idea being examined in the survey...the more likely it is that the respondent's answer is influenced by preexisting impressions, by expectations about what commercials generally say . . . or by guessing, rather than by the actual content of the commercial message or trademark being evaluated." Id. at 256-57. Thus, without a control cell, a likelihood of confusion survey should not survive Daubert. See Simon Prop. Group L.P. v. mySimon, Inc., 104 F. Supp. 2d 1033, 1045-47 (S.D. Ind. 2000); Nat'l Football League Prop., Inc. v. Prostyle, Inc., 57 F. Supp. 2d 665, 668-70 (E.D. Wisc. 1999).

22

13

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 14 of 16

The one-sheet format also introduced an order and location bias by placing "Lulu" across from "Hulu" in columns;23 and · The Failure to Report Data: Although Van Liere indicates that significant information was generated as to the basis for various beliefs (see Van Liere Report, App. C, at 2-3, listing discussion questions), none of that is provided to the Court or Defendant. Rather, selective, self-serving quotes are provided (see id. at 10-12) without the full array. See also

Supplemental Ford Decl. ¶ 16 ("the NERA focus group report is incomplete and does not provide the necessary data and information"). CONCLUSION In the event that this Court grants leave to file Plaintiff's Expert Reports notwithstanding that they are untimely, it should nonetheless find those survey reports and conclusions inadmissible, and should strike the expert reports in their entirety under Daubert. Respectfully submitted, this the 15th day of October, 2007.

/s/ Hayden J. Silver, III Hayden J. Silver, III NC State Bar No. 10037 [e-mail] Betsy Cooke NC State Bar No. 25353 [e-mail] KILPATRICK STOCKTON LLP 3737 Glenwood Avenue, Suite 400 Raleigh, NC 27612 [phone] [phone] (facsimile)
23

"The order in which questions are asked ... and the order in which response alternatives are provided in a closed-ended question can influence the answer." Diamond, "Reference Guide" at p. 254. "To control for order effects, the order of the questions and the order of the response choices in a survey should be rotated . . . ." Id. at 255.

14

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 15 of 16

William H. Brewster GA State Bar No. 080422 [e-mail] Sara Maurer GA State Bar No. 159056 [e-mail] KILPATRICK STOCKTON LLP 1100 Peachtree Street Suite 2800 Atlanta, GA 30309 [phone] [phone] facsimile Attorneys for Defendant Hulu, LLC

15

Case 5:07-cv-00347-D

Document 112

Filed 10/15/2007

Page 16 of 16

CERTIFICATE OF SERVICE This is to certify that on this date the foregoing was electronically filed with the Clerk of the Court using the CM/ECF System which will send notice to the following CM/ECF participants: Leslie C. O'Toole [e-mail] Thomas Hamilton Segars [e-mail] This the 15th day of October, 2007.

/s/ Hayden J. Silver, III

KILPATRICK STOCKTON LLP 3737 Glenwood Avenue, Suite 400 Raleigh, North Carolina 27612 Telephone: [phone]

16