FPX, LLC v. Google, Inc. et al
***DEFICIENT DOCUMENT, PLEASE IGNORE.***MOTION Exclude the Expert Report and Opinion of Thomas J. Maronick and Brief in Support by AOL, LLC., Google, Inc., IAC/InterActiveCorp, MySpace, Inc., Turner Broadcasting System, Inc., YouTube, LLC. (Attachments: # 1 Declaration of Carl Butzer, # 2 Exhibit A to Declaration of Carl Butzer - part 1, # 3 Exhibit A to Declaration of Carl Butzer - part 2, # 4 Exhibit A to Declaration of Carl Butzer - part 3, # 5 Exhibit A to Declaration of Carl Butzer - part 4, # 6 Exhibit A to Declaration of Carl Butzer - part 5, # 7 Exhibit A to Declaration of Carl Butzer - part 6, # 8 Exhibit A to Declaration of Carl Butzer - part 7, # 9 Exhibit A to Declaration of Carl Butzer - part 8, # 10 Exhibit A to Declaration of Carl Butzer - part 9, # 11 Exhibit A to Declaration of Carl Butzer - part 10, # 12 Exhibit A to Declaration of Carl Butzer - part 11, # 13 Exhibit A to Declaration of Carl Butzer - part 12, # 14 Exhibit A to Declaration of Carl Butzer - part 13, # 15 Exhibit A to Declaration of Carl Butzer - part 14, # 16 Exhibit B to Declaration of Carl Butzer, # 17 Declaration of Itamar Simonson, # 18 Exhibit 1 to the Declaration of Itamar Simonson, # 19 Text of Proposed Order)(Babcock, Charles) Modified on 9/20/2010 (sm, ).
FPX, LLC v. Google, Inc. et al
UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF TEXAS MARSHALL DIVISION FPX, LLC d/b/a FIREPOND, Individually and on Behalf of All Others Similarly Situated, Plaintiff, v. GOOGLE, INC., YOUTUBE, LLC, AOL, LLC, TURNER BROADCASTING SYSTEM, INC., MYSPACE, INC. AND IAC/INTERACTIVECORP, Defendants. THE RODNEY A. HAMILTON LIVING TRUST and JOHN BECK AMAZING PROFITS, LLC, Individually and on Behalf of All Others Similarly Situated, Plaintiffs, v. (1) GOOGLE INC.; AND (2) AOL LLC, Defendants. § § § § § § § § § § § § § § § § § § § § § § § § § § § § § §
Civil Action No. 2:09-cv-00142-TJW
CLASS ACTION COMPLAINT
JURY TRIAL REQUESTED
Civil Action No. 2:09-cv-00151-TJW-CE
CLASS ACTION COMPLAINT
JURY TRIAL REQUESTED
DEFENDANTS' CONSOLIDATED MOTION TO EXCLUDE THE EXPERT REPORT AND OPINION OF THOMAS J. MARONICK AND BRIEF IN SUPPORT
Defendants Google Inc. ("Google"), YouTube, LLC, AOL Inc., Turner Broadcasting System, Inc.,1 MySpace, Inc. and IAC/INTERACTIVECORP2 (the "Defendants") respectfully submit this Consolidated Motion to Exclude the Expert Report and Opinion of Thomas J. Maronick and Brief in Support ("Motion to Exclude"). I. INTRODUCTION The two putative trademark class actions pending before this Court demonstrate a fundamental misunderstanding of the trademark laws, which provide trademark owners only with a narrow set of intangible and individualized rights different from other types of intellectual property. "Trade-marks are not monopolistic grants like patents and copyrights." S. Rep. No. 79-1333 (1946), reprinted in 1946 U.S.C.C.A.N. 1274, 1275. The "protections offered by trademark law are less than those accorded copyright or patents. For instance, unlike owners of patents or copyrights, trademark owners do not have exclusive use of a mark." Federal Judicial Center, Manual for Complex Litigation § 33.32, at 640-41 (4th ed. 2004) ("FJC Manual"). Indeed, a trademark "does not confer a right to prohibit the use of the word or words," but rather only "the right to prohibit the use of it so far as to protect the owner's good will against the sale of another's product as his." Prestonettes, Inc. v. Coty, 264 U.S. 359, 369 (1924) (Holmes, J.). As a result, trademark infringement cases focus overwhelmingly on individual factual issues. These class actions are no exception. In order to support a finding of infringement for any given mark, numerous individualized inquiries particular to that specific mark are required. For example, fact intensive inquiries will arise involving (1) issues of ownership and validity, despite the rebuttable presumptions provided by some federal registrations, (2) authorized and unauthorized users, (3) competing rights in different fields of use, (4) affirmative defenses, such
Turner is not a proper defendant and expressly reserves its rights. Based on the allegations in the complaint - which are denied - it appears that one of Turner's subsidiaries, CNN Interactive Group, Inc. is the proper defendant. 2 The proper defendant is IAC/InterActiveCorp's subsidiary, IAC Search & Media, Inc.
Defendants' Consolidated Motion to Exclude Page 1
as abandonment, fraud, fair use, unclean hands, and laches, (5) the multi-factor test for likelihood of confusion, and (6) the availability of damages under the Lanham Act. It is therefore not surprising that no court has ever certified a trademark class action. See, e.g., Vulcan Golf, LLC v. Google Inc., 254 F.R.D. 521, 537 (N.D. Ill. 2008) (denying class certification). Plaint iffs' class certification efforts hinge on the report and two Internet surveys of Professor Thomas J. Maronick (the "Maronick Surveys"). (Mar. Dep., 22-23, Ex.2). Quite simply, the Maronick Surveys suffer from serious methodological flaws, rendering their results both irrelevant and wholly unreliable. Among other things, they failed to: (1) survey the proper universes of relevant consumers; (2) include any controls (and to the extent Maronick claims he did, such "controls" evidenced no confusion); (3) approximate marketplace conditions; and (4) follow any recognized methodology. In short, the surveys violate almost every basic principle and standard for confusion surveys, making their results utterly meaningless. These surveys are also wholly devoid of any relevant information specific to the putative class representatives' own trademarks, they fail to test for commonality across the proposed class members, and they even neglect to consider a means for identifying "all of those sponsored links that in [his] view are initially interest confusing and those for which no consumer would be confused." (Mar. Dep., 194-95). See Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 591 (1993) (noting expert testimony must "fit" the case). It is clear that the Maronick Surveys are "so badly flawed" that any reliance on them would be unreasonable as a matter of law. See Scott Fetzer Co. v. House of Vacuums Inc., 381 F.3d 477, 488 (5th Cir. 2004) (survey "debunked"). The Maronick Surveys should be excluded. II. BACKGROUND A. THE ADWORDS PROGRAM. Google offers its popular search engine services to Internet users, free of charge. To use
Defendants' Consolidated Motion to Exclude Page 2
Google's search engine, users enter a word or phrase query into the search box and press enter. Google then returns a series of search results pages displaying a list of links to websites that Google has algorithmically determined to be relevant to the users' query. 3 In addition to unpaid search results ("organic search results"), Google's search pages often contain a limited number of paid relevant ads ("sponsored links") under the heading "Sponsored Links" above or to the right of the organic search results. Google determines which sponsored links are relevant, in part, by matching words in a users' query with "keywords" selected by advertisers as triggers to have their ads displayed. If an ad is algorithmically determined by Google to relate to a particular user search query, the ad may be displayed along side the organic search results provided by Google in response to such user query. As one court concluded in granting summary judgment for Google on virtually the same causes of action asserted here: "It is in Google's own business interest . . not to confuse its users. . . Google's success depends on its users finding relevant responses to their inquiries." Rosetta Stone Ltd. v. Google Inc., --- F.Supp.2d ----, 2010 WL 3063152, at *8 (E.D. Va. Aug. 3, 2010). B. THE COMPLAINTS. The complaints in these actions allege that the Defendants infringe Plaintiffs' and the putative class members' rights because ads may appear in response to user queries for a word or term that also happens to be a federally registered trademark, even where the advertiser is not the trademark owner. The plaintiffs in the Beck lawsuit propose to represent a class consisting of thousands of U.S.-domiciled federal trademark registrants (Beck Amended Compl. ¶2.a.), and FPX LLC seeks to represent a class of thousands of Texas-domiciled federal trademark registrants in the other suit. (FPX Compl. ¶2.a.).
Google also provides its search services through websites such as AOL, CNN.com, MySpace.com, and ASK.com.
Defendants' Consolidated Motion to Exclude Page 3
III. THE MARONICK SURVEYS The Maronick Surveys in no way address the specific trademarks of any putative class representative,4 a highly unusual omission considering that "as goes the claim of the named plaintiff, so go the claims of the class." Sprague v. General Motors Corp., 133 F.3d 388, 399 (6th Cir. 1998). Instead, Maronick performed two entirely unreliable internet surveys, one involving hypothetical sponsored links ("Survey 1") and the other consisting of two separate surveys ("Survey 2") for the actual trademarks "Southwest Airlines" and "Trek." A. SURVEY 1. Rather than following established survey methodology, Survey 1 presented survey participants with a series of biased and leading questions about search engines and fictional sponsored links that were not actually displayed by Google in response to queries on Google. After prodding participants with questions regarding how frequently they used "a company's name or trademark (e.g., "Nike" or "McDonald's" or "Big Mac")" as a search term (Q.2), and whether they were "familiar with `sponsored links'" which, Maronick coached, are "often at the top or side of the search results page" (Q.4), Maronick asked the following leading questions: · "When you search using a specific company name or trademark, do you expect the Google.com sponsored links to be "sponsored" by that particular company or trademark holder?" (Q.6 in the Google version of Survey 1). (emphasis added).5 · If they would then expect the Google.com sponsored links to be "related to" . . . "in some way" (Q.7) and "affiliated with" (Q.8) the searched-for company or trademark holder. (emphasis added).6 · If they thought they could buy the "searched for" company or trademark holder's goods or services if they clicked on the sponsored link (Q.9). The survey concluded with several irrelevant, tainted questions about Apple's iPad product and two "hypothetical" sponsored links: (1) if the respondents knew "Apple"
The Plaintiffs' complained-of marks are "Firepond," "Firepond CPQ," "Genuinebrands," and "John Beck's." The question is blatantly leading because it provides the "right" answer in the question itself. See infra, Section E. 6 "Related to"... in some way" is the same impermissible prodding found fatal to the "debunked" survey in Scott Fetzer, 381 F.3d at 488, which asked if the defendant was "in any way affiliated with . . ."
Defendants' Consolidated Motion to Exclude Page 4
manufactured the iPad, and (2) whether they thought they would be able to buy the iPad from an "Apple.com" sponsored link (Q.12) and from two other (admittedly fictional) links called "Bob's Electronic Mart" (Q.13) and "iPadd $89.34 On Sale" (Q.14).7 B. SURVEY 2. Survey 2 was actually two separate consumer surveys, covering searches for "Southwest Airlines" ("SWA") tickets and "Trek" bicycles, respectively. By his own admission, Maronick specifically selected these two marks because, in his estimation, they were likely to produce a higher rate of initial interest confusion. Maronick considered, but ultimately rejected, marks like "Neiman Marcus" because of concerns that consumers would not be confused by the sponsored links of competitor department stores, such as "Bloomingdales" and "Bergdorf Goodman." As Maronick testified, links for "Bloomingdales" and "Bergdorf Goodman" relate to "well-known brands, so one wouldn't expect to . . . buy Neiman Marcus clothes at a Bergdorf or Bloomingdales stores. Clearly they were competitive. Searchers were not likely to think, yes, I can buy clothes from Neiman Marcus at Bloomingdales." (Mar. Dep., 92-93).8 In short, Maronick readily admits that certain combinations of marks and sponsored links would "clearly not" produce any likelihood of initial interest confusion, and for that reason purposely chose to avoid testing such marks. Id. 1. Southwest Airlines.
Maronick selected the mark "Southwest Airlines" because he believed at least some confusion would exist due to the unique, individualized trait that "you can't buy Southwest Airlines tickets on anything other than [the] Southwest Airlines [website]." (Mar. Dep., 95). To best leverage this potential for confusion, Maronick omitted asking respondents whether they
The questions do not test actual market conditions, and thus, cannot provide relevant findings. See infra, Section D. "[C]an I buy Dunkin' Donuts at McDonald's or Krispy Kreme donuts. Clearly not." (Mar. Dep., 92).
Defendants' Consolidated Motion to Exclude Page 5
had previously booked SWA tickets online. (Mar. Dep. Ex. 2, p. 64). Obviously, as Maronick admitted, consumers that had previously booked SWA tickets online would be unlikely to be confused because they would already know that "if I want a Southwest ticket I have to go to Southwest Airlines." (Mar. Dep., 59-63). Moreover, Maronick did not take into account
whether survey participants are members of the relevant public, namely, those interested in purchasing airline tickets using a search engine. (Mar. Dep., 271-72); American Footwear Corp. v. General Footwear Co., 609 F.2d 655, 661 n.4 (2d Cir. 1979) (rejecting survey in which participants "did not necessarily have any present purchasing interest."). Participants were then shown only a portion of a Google search results page for the phrase "southwest airlines" (Q.6)9, and directed to focus on the particular sponsored link for "Smartfares.com/Southwest" (Q.7). Maronick again admitted he did not randomly select this particular sponsored link out of the three sponsored links displayed on the page, but rather chose it because it "seemed like one that is more realistically something where they [respondents] think they could buy it [southwest airline ticket]." (Mar. Dep., 280-81). Maronick did not ask
respondents anything about the SWA sponsored link that appeared at the top of the Google search results page because "it says it's Southwest's official site," with the brand name in it. (Mar. Dep., 82-85, 280). Nor did he survey about the link to SWA's actual website that appears as the top organic search result. In other words, his survey focused only on the advertisement most likely, in his predetermined view, to demonstrate likely confusion. After directing the respondents' attention to this one link, which he believed most likely to create confusion, Maronick removed the visual stimulus.10 The remaining questions simply became a memory test in which he asked respondents leading questions about the likelihood that
The abbreviated page included only four organic links and three sponsored links. (Mar. Dep. Ex. 2, p. 65). After the internet survey respondents answered Q.7 they could no longer view the SWA search page results and could not go back to that screen shot for review to assist in answering Q. 8, 9, 10 and 11. (Mar. Dep., 283).
Defendants' Consolidated Motion to Exclude Page 6
respondents could buy a ticket for SWA if they clicked on "the `Smartfares.com/Southwest' sponsored link" (Q.8), and whether the sponsored link is "associated" with (Q.9), "sponsored" by (Q.10), and "affiliated" with SWA (Q.11). (Mar. Dep. Ex. 2, pp. 65-66). 2. Trek.
Maronick chose the mark "Trek," a "lesser well-known brand," at the suggestion of Plaintiffs' attorneys. (Mar. Dep., 198-99, 288-89; Mar. Dep. Ex. 2, p. 7 n.3). Again, selection of the tested mark was not random, but rather selected in the hopes of manufacturing survey results that would support an allegation of likely confusion. Maronick never gauged whether any of the Trek respondents were likely to purchase a new bicycle within any specified time period. (Mar. Dep., 286-87, 292; Mar. Dep. Ex. 2, p. 66-72). Instead, the survey population was limited to people who already owned bicycles (Q.1), rode their bicycles at least seldomly (Q.2), had heard of Trek bicycles (Q.7), and would consider buying a Trek bicycle if they were in the market for a new bicycle (Q.8). But respondents were not asked if they would likely be in the market for a bike. Respondents then were shown only part of a search results page for the search term "Trek" (Q.9), and were told to focus on the one particular sponsored link "`Bikes Up to 60% Off List' from www.BikesDirect.com" (Q.10) out of several sponsored links displayed. After the visual stimulus was removed, Maronick asked the same questions posed in the SWA survey. (Q.12-14). IV. ARGUMENT AND AUTHORITIES In assessing the propriety of a survey, a court must first look to "the manner of conducting the survey," especially the adequacy of the universe; and "second, the way in which participants are questioned." Scott Fetzer Co., 381 F.3d at 487. Because the Maronick Surveys are methodologically flawed, they are irrelevant and unreliable, and should thus be excluded.11
See Paz v. Brush Engineered Materials, Inc., 555 F.3d 383, 388 (5th Cir. 2009) ("[A]ny step that renders the analysis unreliable . . . renders the expert testimony inadmissible.").
Defendants' Consolidated Motion to Exclude Page 7
LEGAL STANDARD TO EXCLUDE EXPERT TESTIMONY. This Court acts as a gatekeeper to ensure that an expert's testimony is both reliable and
Daubert, 509 U.S. at 597.
"In order to consider Plaintiffs' motion for class
certification . . ., the Court must first determine whether Plaintiffs' expert testimony supporting class certification is reliable." Unger v. Amedisys Inc., 401 F.3d 316, 323 n.6 (5th Cir. 2005).12 Surveys suffering from serious flaws such as those discussed herein require exclusion. See, e.g., Scott Fetzer Co., 381 F.3d at 487-88.13 Plaintiffs cannot establish their burden of proving that Daubert's "exacting standards" are met. Weisgram v. Marley Co., 528 U.S. 440, 455 (2000). B. THE MARONICK SURVEYS FAILED TO TARGET THE RELEVANT CONSUMER UNIVERSES. The "adequacy of the `survey universe'" is one of the "most important" factors in assessing the validity of a survey. Amstar Corp. v. Domino's Pizza, Inc., 615 F.2d 252, 264 (5th Cir. 1980). "Selection of the proper universe is a crucial step, for even if the proper questions are asked in a proper manner, if the wrong persons are asked, the results are likely to be irrelevant."14 None of the Maronick Surveys targeted the relevant universe of likely purchasers. The universe for Survey 1 is fatally flawed because Maronick failed to ask respondents whether they were likely to purchase any goods or services. Scott Fetzer Co., 381 F.3d at 48788. Maronick did not ask if respondents would consider buying an iPad/tablet computer (or even whether they would use the Google search engine). (Simonson ¶7615; Mar. Dep., Ex. 2, 65-66).
See also Bell v. Ascendant Solutions, 422 F.3d 307, 314 n.13 (5th Cir. 2005) (finding it appropriate to consider the reliability of expert testimony "at the class certification stage."); Brooks v. Lincoln Nat'l Life Ins. Co., No. 5:03CV-256, 2008 WL 4355390 (E.D. Tex. Feb. 12, 2008) (excluding expert at class certification stage) (Craven, M.J.). 13 See also Citizens Fin. Group, Inc. v. Citizens Nat'l Bank, 383 F.3d 110 (3d Cir. 2004); THOIP v. Walt Disney Co., 690 F. Supp. 2d 218, 233 (S.D.N.Y. 2010); Wells Fargo & Co. v. WhenU.com, Inc., 293 F. Supp.2d 734, 767 (E.D. Mich. 2003); Simon Prop. Group L.P. v. mySimon, Inc., 104 F. Supp. 2d 1033, 1052 (S.D. Ind. 2000). 14 6 J. Thomas McCarthy, McCarthy on Trademarks & Unfair Competition, §32.159 (4th ed. 2010) ["McCarthy"]; Shari Seidman Diamond, "Reference Guide on Survey Research," in Fed. Jud. Ctr., Reference Manual On Scientific Evidence, 229, 239, 240 (2d ed. 2000) ["Diamond"]. Maronick recognizes McCarthy and Diamond as "authoritative." (Mar. Dep., 12-15). 15 "Simonson" refers to Dr. Itamar Simonson's expert report attached to his Declaration as Exhibit "1."
Defendants' Consolidated Motion to Exclude Page 8
The universes for Survey 2 are likewise improper. The SWA survey failed to ask respondents if they expected to purchase an online flight reservation within a specified period (e.g., six months) and, if so, whether they intended to use a search engine for that purpose. (Mar. Dep., 131-32, 272; Simonson ¶76). In the Trek survey, Maronick improperly excluded persons who did not own a bicycle. (Mar. Dep., 147, 149, 286-89); See Scott Fetzer Co., 381 F.3d at 487 (excluding a survey that failed to include consumers who did not own a vacuum cleaner). Compounding his error, Maronick asked the remaining, bike-owner participants to confirm that they (1) had heard of Trek bicycles, and (2) would "consider buying a Trek bicycle if you were in the market for a new bicycle?" (Mar. Dep. Ex. 2, p. 80 (emphasis added)). Maronick even admits he never qualified the Trek participants as prospective purchasers. (Mar. Dep., 292; Simonson ¶75). Because Maronick failed to target the proper universes of consumers, his surveys are not relevant and must be excluded. See Simonson ¶74; Scott Fetzer Co., 381 F.3d at 487.16 C. THE MARONICK SURVEYS FAILED TO USE ANY CONTROLS. A survey designed to estimate likelihood of confusion that fails to include a proper control suffers from a major flaw. THOIP, 690 F. Supp. 2d at 240. Absent proper control, there is no benchmark for determining whether a likelihood of confusion estimate is significant or merely reflects guessing and the flaws of the survey methodology. Id.; Diamo nd at 260.17 The Maronick Surveys failed to use any controls and can be excluded on this basis alone. Alt hough he readily admitted he did not use a control in Survey 2 for either SWA or Trek (Mar. Dep., 216-17), Maronick initially testified that Survey 1 included two controls. First, he
See also American Footwear Corp., 609 F.2d at 661 n.4 (rejecting survey in which participants did not have a present purchasing interest); Amstar Corp., 615 F.2d at 263-64 (surveys substantially defective). 17 Control questions are used to measure the origins of the participants' perceptions about products in order to assure that the participants are not basing their answers on preconceptions and also provide the ability to evaluate the effect of the wording of a particular question. FJC Manual at 256, 258.
Defendants' Consolidated Motion to Exclude Page 9
contended the questions and results for the search engine surveys operated as a control for the remaining portions of the survey, and that this "serves as an overall control for the whole study, because ... there's no stimuli for comparison." (Mar. Dep., 216-17, 220-21). He also testified "the use of the apple.com sponsored link served to a certain extent as a control because consumers know that the iPad is made by Apple, so it served a function of a control." (Mar. Dep., 217). His testimony in this regard is incoherent, contradictory and suggests that he does not understand the basic concept of what a control is and the manner in which it should be used. (Simonson ¶70). Even assuming Maronick were correct about what constitutes a control (he is not), Survey 1 should have led him to conclude that there was no confusion, since the net estimate for likelihood of confusion is calculated by subtracting the confusion estimate in the control group from the estimate in the "test" group. THOIP, 690 F. Supp.2d at 240. Doing so here, based on Maronick's Apple.com "control" results, the net estimate in the present case was negative; that is, the "confusion" in the Apple.com control group (83%) was greater than in the test group (53%) leading to a net confusion of negative 30%. (Simonson ¶32). In other words, no confusion. Realizing the absurdity of this position, Maronick ultimately admitted, at the end of his deposition, that he did not use a control in Survey 1. (Mar. Dep., 311-12). In sum, Maronick's report and his surveys should be excluded because the lack of a control group renders the survey results fatally flawed and meaningless.18 D. THE MARONICK SURVEYS FAILED TO REPLICATE MARKETPLACE CONDITIONS. The failure of the Maronick Surveys to replicate actual marketplace conditions warrant their exclusion. See American Footwear Corp., 609 F.2d at 661 n.4. In Survey 1, Maronick made up his own hypothetical sponsored links. (Mar. Dep., 257-58). In Survey 2, respondents
THOIP, 690 F. Supp. 2d at 240 (excluding report); Wells Fargo & Co., 293 F. Supp.2d at 768-69.
Defendants' Consolidated Motion to Exclude Page 10
were shown portions of static, non-interactive search results pages containing sponsored links. (Mar. Dep., 152, 277-79; Mar. Dep. Ex. 2, pp. 65, 70). Such survey stimuli do not accurately reflect what a consumer would actually encounter when using Google's search services, and thus results based thereon should be excluded. THOIP, 690 F. Supp.2d at 231. This is particularly true when, as is the case here, the survey is "questioning interviewees about a web page image that differs from the actual web page in issue." McCarthy, §32.171.19 Respondents were also not asked the relevant questions upon which Maronick's conclusions were drawn until after the stimuli were removed from view. (Mar. Dep. Ex. 2, 6566, 70). When the survey respondents were asked whether they could purchase an SWA ticket from the particular sponsored link chosen by Maronick, they could no longer view any sponsored links or "organic" listings, including the first sponsored link for SWA and the first "organic" listing on the SWA results page that stated: "Official Southwest Airlines website; the only place to find Southwest Airlines fares online." (Id. at 65-66). This scenario is unlike the marketplace conditions Google users would actually encounter. Maronick's failure to allow respondents to view actual search engine results is a critical alteration that turns the survey into "nothing more than a meaningless memory game . . . that bears no relationship to the marketplace." Simon Prop. Group, 104 F. Supp.2d at 1043. Moreover, the use of a partial, static page followed by the removal of the stimuli altogether "removes the inevitable need for some effort on the part of the Internet user to sort through the responses" and "distort[s] the experience by removing the additional information available to help sort through those results." Id. at 1044. The obvious effect is to exaggerate any confusion that might be detected, which thoroughly undermines the reliability of the surveys. Id.
See also Kargo Global, Inc. v. Advance Magazine Pub., Inc., No. 06 Civ. 550, 2007 WL 2258688 at *10 (S.D.N.Y. Aug. 6, 2007); Wells Fargo & Co., 293 F. Supp.2d at 765.
Defendants' Consolidated Motion to Exclude Page 11
Maronick's failure to replicate market conditions is fatal and his opinions should be excluded. E. THE MARONICK SURVEYS FAILED TO FOLLOW ANY RECOGNIZED METHODOLOGY. The Maronick Surveys failed to employ any accepted or recognized methodology for testing initial interest confusion. (Simonson ¶¶56-57).20 Maronick's "methodology" consisted of biased, leading statements that essentially asked respondents to confirm that they agreed. (Simonson ¶57). For that reason alone, the results are unreliable and should be excluded.21 1. The Maronick Surveys were Designed to Produce Pre-determined Results.
In consumer surveys, it is improper to use leading questions that assert a business relationship when the respondent may not have thought that such a connection existed. (Simonson ¶46). When designing a survey, the researcher must also avoid what is known as "demand effect" and "order effect" bias. (Id. ¶¶46-47). A "demand effect" results when the questions or other elements of the survey design influence participants' responses by suggesting what the "correct" answers might be or by implying associations that might not otherwise occur to participants. Id.; see Simon Prop. Group, 104 F. Supp.2d at 1048 (excluded); American Footwear Corp., 609 F.2d at 661 n.4 (same). An "order effect" results when a participant's answer to one question affects his answers to subsequent questions. Id. ¶61; Diamond at 274. Both effects can significantly bias the survey results. Id. Maronick posed leading, demand effect questions throughout the surveys. For example, the first pertinent question in Survey 1 was Q.6: "When you search using a specific company
Most methodologies can be divided into two categories: (1) methods in which respondents are shown just one of the marks and asked to identify the company that puts it out (the Eveready format is the most common method), or (2) methods in which respondents are shown both the junior and senior marks and asked questions to assess confusion (e.g., the "line-up survey" method and the Squirt "side-by-side" method). (Simonson ¶¶41-43). According to McCarthy, the Eveready methodology is the standard survey format, McCarthy, §32:174, although Maronick conceded he could not even explain what an Eveready survey is. (Mar. Dep., 302-03; 307). 21 See Wells Fargo & Co., 293 F. Supp.2d at 753 (survey deemed unreliable because expert used leading questions that may have skewed the survey results); Vail Assoc., Inc. v. Vend-Tel-Co., Ltd., 516 F.3d 853, 864 n.8 (10th Cir. 2008) (exclusion due to, among other things, question bias and participant bias).
Defendants' Consolidated Motion to Exclude Page 12
name or trademark, do you expect the Google.com sponsored links to be "sponsored" by that particular company or trademark holder?" (Emphasis added).22 The obvious flaw is that the question includes the word "sponsored" twice, but it does not define the meaning of each "sponsored."23 This suffers from strong "demand effects" whereby respondents know the
"correct" answer and what they are supposed to say. (Simonson ¶59a); see Scott Fetzer, 381 F.3d at 488 (a survey question that "begs its answer ... cannot be a true indicator of the likelihood of consumer confusion."). The question also uses biased wording to capitalize on the "acquiescence bias," whereby respondents are significantly more likely to agree with provided statements than to disagree. (Simonson ¶59b). In the face of such a glaringly biased question, Maronick admitted "[t]here is a problem with the fact that the sponsored is in there twice..." (Mar. Dep., 243), and he "would have been surprised" if the survey participants had answered anything other than that a sponsored link was "sponsored." (Id., 240-42). The remaining questions in Survey 1 suffer from similar problems, but with one additional flaw. The questions asked after Question 6 created an "order effect" bias because respondents typically try to provide answers that do not contradict earlier answers. (Simonson ¶61). In other words, once respondents agreed that a sponsored link is "sponsored" with the company/trademark, it would make little sense for them to disagree with the subsequent statements that the sponsored links were also "related" (Q.7) and "affiliated" (Q.8). (Id.). 2. The Maronick Surveys Relied on Closed-Ended Questions and Failed to Ask Respondents to Explain Their Answers.
Maronick relied exclusively on closed-ended questions whereby respondents were given biased answers, which in all cases corresponded to the Plaintiffs' position. (Simonson ¶72).
Respondents were not shown any visual stimuli. (Mar. Dep. Ex. 2, 35-38). The Maronick Surveys failed to explain the terms "related," "affiliated," "sponsored" and "associated." See Mary Kay, Inc. v. Weber, 601 F. Supp.2d 839, 848 (N.D. Tex. 2009) (criticizing expert's failure to define "affiliation").
Defendants' Consolidated Motion to Exclude Page 13
Unlike the Maronick Surveys, sound confusion surveys generally include open-ended questions seeking explanations to answers given by respondents. Id. As Professor McCarthy observes: Both questions should be followed up by the important question: `What makes you say that?' Often, an examination of the respondents' verbatim responses to the `why' question are the most illuminating and probative part of the survey, for they provide a window into consumer thought processes in a way that mere statistical data cannot. McCarthy, §32.175. Maronick did not ask this crucial follow-up question, instead choosing to leave unexplored the reasons behind the responses to his surveys' biased, leading questions.24 In sum, Maronick's failure to follow a recognized methodology in either survey serves as proof that a flawed survey can produce almost any result. (Simonson ¶37). F. THE MARONICK SURVEYS ARE IRRELEVANT TO THE CLASS INQUIRY. Pursuant to Rule 702, when experts opine on issues irrelevant to the case, courts must exclude their testimony. Daubert, 509 U.S. at 591. In Daubert, the court described the
connection between relevancy of testimony and issues in the case at hand as one of "fit," noting that "`[f]it' is not always obvious, and scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes." Daubert, 509 U.S. at 591. In the context of traditional Lanham Act claims, "fit" requires that an expert conducting a confusion survey target the relevant question of confusion. Scotts Co. v. United Indus. Corp., 315 F.3d 264, 278-80 (4th Cir. 2002) (abuse of discretion to credit a survey that failed to establish consumer confusion on a relevant question).25 The Maronick Surveys have nothing to do with Plaintiffs' own trademarks, (Mar. Dep., 70), or even the industries in which they operate; nor do the surveys even say much about the "Southwest Airlines" or "Trek" trademarks. More importantly, the flawed Maronick Surveys do not speak to whether classes of tens
See Diamond at 260 ("[T]he response alternatives in a closed ended question may remind respondents of options that they would not otherwise consider or which simply do not come to mind as easily"). When asked about including open-ended questions, Maronick stated that "it didn't seem that there was a need to." (Mar. Dep., 248-49). 25 Mary Kay, 601 F. Supp.2d at 849 (striking expert's survey where results failed to weed out irrelevant confusion).
Defendants' Consolidated Motion to Exclude Page 14
or hundreds of thousands of registered trademark owners should be certified. Defendants' expert, Professor Simonson concluded that for reasons that are grounded in basic principles of consumer behavior and marketing, Plaintiffs' contention that allegations could possibly apply across all of the proposed class members and sponsored links is clearly erroneous for numerous reasons. For example, it is indisputable that trademarks vary in terms of consumer recognition, or that sponsored links vary in content and context, or that there are large differences in consumer experience and care depending upon the goods at issue. (Simonson ¶11). As Maronick himself admits: (1) his survey results do not suggest that all sponsored links cause initial interest confusion, (Mar. Dep., 193), (2) he has not attempted to identify those sponsored links "that might be initially interest confusing from those that are not confusing," (Mar. Dep., 194-95, 205, 208), (3) if he did attempt such identification the results would vary depending on strength of brand and the content of the corresponding sponsored links, (Mar. Dep., 198, 203), and (4) "identifying the perceived relationship between `searched for' trademarks and sponsored links is likely to be exponentially more difficult and time consuming with less well known brands." (Mar. Dep. Ex. 2, p. 7 n.3 (emphasis added)). Thus, the Maronick Surveys do not provide any pertinent information regarding the other proposed class members, and in particular any shared commonality or superiority of the putative classes, and thus are irrelevant. (Simonson ¶12).26 V. PRAYER Defendants respectfully request that their Consolidated Motion to Exclude the Expert Report and Opinion of Thomas J. Maronick and Brief in Support be granted.
The Court also should strike Maronick's conclusory and untested assertion, (Mar. Dep. Ex. 2, p. 20), that "Google has created an expectation that consumers will be able to buy the company's products or services from the sponsored link firm by the inclusion of the sponsored link on the `searched for' company or trademark results page." See Hathaway v. Bazany, 507 F.3d 312, 318-19 (5th Cir. 2007) (court properly excluded testimony, because it relied on insufficient factual support and "a host of unsupported conjectures that falls far short of a methodology").
Defendants' Consolidated Motion to Exclude Page 15
Respect fully submitted, /s/ Charles L. Babcock CHARLES L. "CHIP" BABCOCK Texas State Bar No. 01479500 JACKSON WALKER L.L.P. 901 Main Street, Suite 6000 Dallas, Texas 75202 (214) 953-6030 (214) 953-5822 - Fax Email: email@example.com DAVID T. MORAN Texas State Bar No. 14419400 JACKSON WALKER L.L.P. 901 Main Street, Suite 6000 Dallas, Texas 75202 (214) 953-6051 (214) 661-6677 - Fax Email: firstname.lastname@example.org CARL C. BUTZER Texas State Bar No. 03545900 JACKSON WALKER L.L.P. 901 Main Street, Suite 6000 Dallas, Texas 75202 (214) 953-5902 (214) 661-6609 - Fax Email: email@example.com ATTORNEYS FOR DEFENDANTS
Defendants' Consolidated Motion to Exclude Page 16
CERTIFICATE OF SERVICE I hereby certify that on September 17, 2010, I electronically submitted the foregoing document with the clerk of the court for the U.S. District Court, Eastern District of Texas, using the electronic case files system of the court. The electronic case files system sent a "Notice of Electronic Filing" to individuals who have consented in writing to accept this Notice as service of this document by electronic means. All other counsel of record not deemed to have consented to electronic service were served with a true and correct copy of the foregoing by first class mail today, September 17, 2010.
/s/ Charles L. Babcock Charles L. Babcock
Defendants' Consolidated Motion to Exclude Page 17