IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE
INKIT, INC., Plaintiff, Civil Action No. 23-793-RGA Vv. AIRSLATE, INC., Defendant.
MEMORANDUM OPINION Gabriela Monasterio, Travis Steven Hunter, Jason James Rawnsley, RICHARDS, LAYTON & FINGER, P.A., Wilmington, DE; Sarah Theresa Andrade, BAYARD, P.A., Wilmington, DE; Charles Brandon Browning, H. William Bloom, III, MAYNARD NEXSEN PC, Birmingham, AL, Attorneys for Plaintiff. Mary Claire McCudden, MARSHALL DENNEHEY, PC, Wilmington, DE; Gregor Hensrude, Petra N. Ambrose, KLINEDINST PC, Seattle, WA; Kevin Gramling, KLINEDINST PC, Irvine, CA, Attorneys for Defendant.
March a0 , 2025
Before me are Defendant airSlate’s Daubert motions to exclude the testimony of two of Plaintiff Inkit’s experts, Mr. Glenn Newman and Dr. Justin Anderson. (D.I. 156, 159). Ihave reviewed the parties’ briefing. (D.J. 157, 160, 164, 167, 173, 175). For the reasons set forth below, Defendant’s Daubert motion to exclude the testimony of Mr. Newman is DENIED. Defendant’s Daubert motion to exclude the testimony of Dr. Anderson is GRANTED in part and DENIED in part. I. BACKGROUND Inkit filed the instant suit on July 21, 2023. (D.I. 1). After recently dropping its Lanham Act claim, Inkit now alleges four counts—three state law claims that are similar to various provisions of the Lanham Act, and a breach of contract claim. (D.I. 2 at 13-16; see D.J. 194 at 3 of 25). Previously, on March 2, 2023, Inkit sued airSlate for improper use of Inkit’s trademarks in advertisements. (D.I. 103-1 at 5-8); see Inkit, Inc. v. airSlate, Inc., No. 23-cv-00232 (D. Del. Mar. 2, 2023), D.I. 1. The parties settled the earlier case. The settlement agreement states, “airSlate shall not use or publicly display or cause others to use or publicly display” Inkit’s trademark “in any form or manner, including without limitation in airSlate’s web-based or other advertising, that infringes upon or otherwise violates Inkit’s rights in the Mark[.]” (D.I. 3 at 2 of 6, § 2(a)). Inkit brought the instant suit based on airSlate’s continued display of the word “inkit” on seven of airSlate’s webpages. (D.I. 2 at 4-11). On these webpages, Inkit’s trademark appears alongside the trademark DocuSign.' (/d. at 4] 19). The webpages advertise airSlate’s signNow product as an alternative to “inkit (Docusign).” (/d. at 5-11).
1 The owner of the DocuSign trademark, DocuSign, Inc., is not involved in this case.
airSlate now moves to exclude the testimony of two of Inkit’s experts, Glenn Newman and Dr. Justin Anderson, under Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). (D.I. 156; D.I. 159). II. LEGAL STANDARD Federal Rule of Evidence 702 sets out the requirements for expert witness testimony and states: A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that: (a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case. R. Evip. 702. The Third Circuit has explained: Rule 702 embodies a trilogy of restrictions on expert testimony: qualification, reliability and fit. Qualification refers to the requirement that the witness possess specialized expertise. We have interpreted this requirement liberally, holding that “a broad range of knowledge, skills, and training qualify an expert.” Secondly, the testimony must be reliable; it “must be based on the ‘methods and procedures of science’ rather than on ‘subjective belief or unsupported speculation’; the expert must have ‘good grounds’ for his o[r] her belief. In sum, Daubert holds that an inquiry into the reliability of scientific evidence under Rule 702 requires a determination as to its scientific validity.” Finally, Rule 702 requires that the expert testimony must fit the issues in the case. In other words, the expert’s testimony must be relevant for the purposes of the case and must assist the trier of fact. The Supreme Court explained in Daubert that “Rule 702’s ‘helpfulness’ standard requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility.” Schneider ex rel. Estate of Schneider v. Fried, 320 F.3d 396, 404-05 (3d Cir. 2003) (footnote and internal citations omitted).”
? The Court of Appeals wrote under an earlier version of Rule 702. Subsequent amendments do not alter the applicability of the quoted discussion.
Ili. DISCUSSION Mr. Newman provided an opinion solely about Inkit’s potential recovery in the form of disgorgement. (D.I. 158-4 at 12 of 31 432). Dr. Anderson designed a survey to measure consumer confusion about the relationship between Inkit and DocuSign caused by airSlate’s webpages. (D.I. 161-3 at 4 of 320 3-5). airSlate does not dispute the “qualification” of either Mr. Newman or Dr. Anderson but rather focuses on the “reliability” and “fit” of their respective expert opinions. See Schneider, 320 F.3d at 404. “[I]n order for expert testimony to be reliable, and therefore admissible, it must be based on the methods and procedures of science rather than subjective belief or speculation.” In re TMI Litig., 193 F.3d 613, 670 (3d Cir. 1999), amended by 199 F.3d 158 (3d Cir. 2000). “The test of admissibility is not whether a particular scientific opinion has the best foundation, or even whether the opinion is supported by best methodology or unassailable research. Instead, the court looks to whether the expert’s testimony is supported by ‘good grounds.’” Karlo v. Pittsburgh Glass Works, LLC, 849 F.3d 61, 81 (3d Cir. 2017) (citing Jn re TMI Litig., 193 F.3d at 665). For expert testimony to “fit the issues in the case,” the testimony “must be relevant for the purposes of the case and must assist the trier of fact.” Schneider, 320 F.3d at 404. A. Mr. Glenn Newman On the question of reliability, airSlate raises a number of complaints about Mr. Newman’s expert report, largely concerning items airSlate alleges he did not sufficiently address or review. airSlate claims Mr. Newman did not perform a “proper investigation” because, first, he thought the seven webpages were “examples of purported infringing activity” instead of the totality of that activity, and second, he did not review the deposition testimony of Inkit’s Rule
30(b)(6) representative.? (D.I. 157 at 3; D.J. 158-1 at 10, 12-13 of 88). airSlate’s remaining contentions deal with Mr. Newman’s failure to deduct costs from airSlate’s revenues or to apportion airSlate’s revenue earned solely due to unlawful activity when determining disgorgement. (D.I. 157 at 3—8, 11-16). For example, Mr. Newman reviewed a sample of Google Analytics user data for the webpages purportedly showing there were “no conversions* for all visits to the Seven Webpages,” but declined to rely upon the user data for his opinion. (/d. at 6; D.I. 158-1 at 59 of 88; D.J. 158-5 at 4 of 91, Ex. F). airSlate also contends Mr. Newman should have “whittled down” revenues by, for instance, calculating “proportionate revenues” to account for airSlate’s signNow revenues attributable to similar webpage ads featuring the trademarks of competitors other than Inkit. (D.I. 157 at 11). Inkit points out that Mr. Newman “focused his analysis on sales of airSlate’s signNow product” using “airSlate’s own sales schedules.” (D.I. 164 at 4; D.I. 158-4 at 11, 17-18 of 31 9 28, 44-46; D.I. 166-1 at 34 of 44, Ex. B). Mr. Newman assumed in his report that the entirety of “the monthly sales summary produced by airSlate represent[ed] unlawful sales subject to disgorgement” because airSlate “provided limited information concerning revenues associated with its signNow product.” (D.I. 158-4 at 11 of 31 § 28). Mr. Newman confined his calculations of revenue to three alternative time periods: May 1, 2023 through July 31, 2023;
3 As Inkit points out, Mr. Newman was tasked with showing airSlate’s profits. Inkit’s Rule 30(b)(6) representative could not provide this information. (D.I. 164 at 10). 4 According to Defendant, “A conversion consists of signing up for a free trial or subscribing to the SignNow service” from the associated webpage. (D.I. 157 at 1). > airSlate agrees that it has not provided Inkit with customer lists or customer-specific sales data and notes that I previously denied Inkit’s request for such data. (D.I. 157 at 14; D.L. 93 at 13 of 18).
March 2, 2023 through July 31, 2023; and August 8, 2022 through July 31, 2023.° (d. at 18 of 31 46). As Mr. Newman said in his deposition, “[I]n general, I think it is not proper to include a hundred percent of all of a company’s revenues, in the event that one aspect may have been unlawfully utilized.” (D.I. 158-1 at 27 of 88). However, in undertaking his assignment, Mr. Newman understood that: [T]he Plaintiff has to establish, based upon the records that are available to the Plaintiff, those numbers that potentially is the starting point for the unjust revenues and then the Defendant has—presumably, they have control of their own records, I don’t have their records, to try to whittle it down to those sales that are unjust or at least a portion of the sales that represent unlawful sales. (id. at 32-33 of 88). Mr. Newman’s opinions are “supported by good grounds” and are “based on the methods and procedures of science, not on subjective belief and unsupported speculation.” Karlo, 849 F.3d at 80-81 (citation omitted). Therefore, I] find that Mr. Newman’s opinions are reliable. Mr. Newman was retained by Inkit to provide an expert opinion on Inkit’s potential “monetary recovery solely from an unjust enrichment (i.e. disgorgement of
6 These date range start dates correspond to (1) the effective date of the settlement agreement, (2) the commencement date of the first lawsuit, and (3) the beginning of airSlate’s ad campaign featuring the webpages. (D.I. 158-4 at 13 of 31 34-35). Despite the release of claims in the settlement agreement, Mr. Newman posited that the appropriate infringement period may extend back to the beginning of airSlate’s marketing campaign in August 2022 because airSlate “failed to implement the first provision of the Settlement Agreement (related to the use or display of the Mark) until at least July 21, 2023.” (id. at 15 of 31 939). Regarding the end date of the time periods, Mr. Newman assumed violations would have been cured shortly after the date the complaint was filed in the instant suit: July 21, 2023. (Ud. at 16 of 31 41). Accordingly, he limited the period of recovery to dates prior to July 31, 2023, but “reserve[d] the right to extend such period as appropriate[.]” (/d.).
defendant’s profits) perspective.” (D.I. 158-4 at 12 of 31 932). The Lanham Act? provides that successful plaintiffs may, “subject to the principles of equity,” “recover (1) defendant’s profits, (2) any damages sustained by the plaintiff, and (3) the costs of the action.” 15 U.S.C. § 1117(a). “In assessing profits the plaintiff shall be required to prove defendant’s sales only; defendant must prove all elements of cost or deduction claimed.” Id. The Third Circuit has stated that a plaintiff may satisfy its burden of proof when estimating a defendant’s profits by providing the total sales of an infringing product during the relevant time period. See Banjo Buddies, Inc. v. Renosky, 399 F.3d 168, 176 (3d Cir. 2005). In the event of a finding of infringement, Inkit bears the burden of proving airSlate’s sales stemming from infringement—in this case, revenues derived from airSlate’s signNow product, which is advertised on the seven webpages. airSlate, not Inkit, bears the burden of proving “all elements of cost or deduction,” including which of those sales were not attributable to its unlawful conduct. § 1117(a). Further, as disgorgement is an equitable remedy, the Lanham Act allows the court discretion to “enter judgment for such sum as the court shall find to be just, according to the circumstances of the case.” Jd.
7 During the pre-trial conference, Inkit confirmed its dismissal of the Lanham Act claim. (D.I. 194 at 3 of 25). Disgorgement, or an accounting of a defendant’s profits, is a common law trademark infringement remedy and governed by “analogous” rules “to those applicable under the Lanham Act.” RESTATEMENT (THIRD) OF UNFAIR COMPETITION § 37 cmt. a (AM. L. INST. 1995); see, e.g., Coca-Cola Co. v. Nehi, 25 A.2d 364, 367 (Del. Ch. 1942), decree sustained by 36 A.2d 156 (Del. 1944) (plaintiff seeking “an accounting of damages claimed to have been sustained by it for the illegal profits realized by defendant”). As the parties briefed this issue before the dismissal of the claim, the briefing focuses on the Lanham Act and associated caselaw. I discuss the issue of disgorgement in these terms recognizing that the issue will now be decided under Delaware law.
Newman limited his calculations of sales to three different potential time periods for infringement reasonably bounded by relevant events. As the webpages are advertisements for airSlate’s signNow product, Mr. Newman accordingly limited his calculations of airSlate’s revenues, or “sales,” to those deriving from the signNow product. He derived these sales numbers from the limited information provided to him by airSlate. He also acknowledged in his deposition that a measure of total revenue is not an appropriate measure for disgorgement if only certain elements or sales are unlawful. (D.I. 158-1 at 27 of 88). He likewise acknowledged the assumptions he needed to make in his expert report “[d]Jue to lack of detailed sales data.” (D.I. 158-4 at 17 of 31 43). airSlate cannot refuse to hand over information and then fault Inkit’s expert for failing to use that information in his report to “whittle down the data any further.” (D.I. 157 at 14). Regarding the question of “fit,” expert testimony that “‘fit[s] the issues in the case
... must be relevant for the purposes of the case and must assist the trier of fact.” Schneider, 320 F.3d at 404. Mr. Newman’s testimony about airSlate’s revenues during reasonable time periods based on the sales schedule provided by airSlate is both relevant to the issue of defendant’s unlawful sales and helpful to the trier of fact. airSlate may argue for appropriate limits and deductions on those revenues through cross examination or through its own witnesses. airSlate’s motion to exclude Mr. Newman’s testimony is DENIED. B. Dr. Justin Anderson Generally, a survey’s “technical unreliability goes to the weight accorded a survey, not its admissibility.” Citizens Fin. Grp., Inc. v. Citizens Nat’l Bank, 383 F.3d 110, 121 Gd Cir. 2004). “Typically, a court will not exclude a survey unless it is so flawed that it would be completely
unhelpful or harmful to the trier of fact.” Merisant Co. v. McNeil Nutritionals, LLC, 242 F.R.D. 315, 320 (E.D. Pa. 2007). Dr. Anderson was retained by Inkit to conduct a consumer survey and interpret the results. airSlate moves to exclude Dr. Anderson’s study and opinions as unreliable, arguing primarily: (1) the universe of participants in the survey was underinclusive; (2) the survey designer (Dr. Anderson) was aware of the litigation and the position that was favorable to Inkit; and (3) certain survey questions were inappropriately leading and Dr. Anderson only relied on the answers to these leading questions when formulating his conclusions. Because of these alleged flaws in methodology, airSlate argues that Dr. Anderson’s survey and opinions are not helpful to the trier of fact and therefore also do not satisfy the “fit” requirement. 1. Underinclusive universe “A universe is that segment of the population whose perceptions and state of mind are relevant to the issues in the case. A survey of the wrong universe will be of little probative value in litigation.” Citizens Fin. Grp., 383 F.3d at 118-19 (internal citations and quotation marks omitted). Prospective respondents to Dr. Anderson’s survey were recruited through a survey panel operated by Prodege, “a highly reputable sample provider for the marketing research industry.” (D.I. 161-3 at 16 of 320 § 36). Dr. Anderson limited the participants of the survey to what he regarded as “the relevant consumer population at issue in this matter, which is consumers who purchase document e-signature services as part of their job.” (/d. at 16 of 320 J 35). The only prospective respondents invited to participate were those that indicated to Prodege that they were employed full time, employed part time, or self-employed. (dd. at 17 of 320 § 40(i)). From this sample, only respondents who indicated that they “purchase or select document e-signature
services” as part of their job were included in the survey population and allowed to answer the subsequent questions. (/d. at 17 of 320 § 40(ii)). Dr. Anderson stated that his understanding was that “it’s typically a business that is purchasing . . . [d]ocument e-signature services.” His understanding came from discussions with Inkit’s counsel and his review of the sources listed in his report, including airSlate’s SignNow website and the webpages at issue. (D.I. 161-1 at 14 of 24; see D.I. 161-3 at 7 of 320 { 16). Accordingly, airSlate argues, “Dr. Anderson had no independent understanding of the types of customers airSlate has for its SignNow product, and had no such understanding as to whether those customers were solely businesses.” (D.I. 160 at 5). airSlate also points to statements from its CEO that “the vast majority” of airSlate’s customers are “individuals.” (D.I. 160 at 9; D.I. 161-5 at 3 of 4 { 6). Inkit argues, “[A] survey that is simply underinclusive or has other flaws in sampling remains admissible but subject to challenge.” (D.I. 167 at 6 (citing In re Suboxone (Buprenorphine Hydrochloride & Naloxone) Antitrust Litig., 2020 WL 6887885, at *17 (E.D. Pa. Nov. 24, 2020))). Inkit also denies that the universe is underinclusive, pointing to Dr. Anderson’s testimony about airSlate’s website, which offers “four subscription plans” called “business,” “business premium,” “enterprise which usually refers to a large scale business organization,” and “site license,” which Dr. Anderson characterized as “part of business subscriptions” in his experience. Dr. Anderson went on to say: “[I]t looks to me like these are four different types of business subscription plans.” (D.I. 167 at 8-9; D.I. 168-1 at 376-77 of 417, Ex. D). Further, Inkit argues that “it is unclear . . . how airSlate differentiates an ‘individual’ from those who select such services for their job[.]” (D.I. 167 at 8).
airSlate also argues that the sample was underinclusive because Dr. Anderson’s qualifying questions excluded users who indicated they purchased “document storage services” for their job (and not “document e-signature services”) even though “document storage services” “is expressly included as a service airSlate’s signNow product provides.” (D.I. 160 at 6). In his deposition, Dr. Anderson indicated he regarded the services offered under SignNow “collectively ... aS document e-signature services.” Dr. Anderson assumed that the customers who thought they purchased “document storage services” but not “document e-signature services” would be “thinking of that document storage service differently than the type of services that is offered by SignNow.” (D.I. 161-2 at 18 of 54). Inkit notes that Dr. Anderson regarded the appropriate “consumer population” as “potential purchasers of document e-signature services, and not specifically purchasers, actual purchasers of airSlate’s SignNow service.” (D.I. 167 at 10 (citing D.I. 168-1 at 377 of 417, Ex. D)). Inkit argues, “Just because airSlate offers document storage services does not mean that SignNow does, and does not mean that it is a part of the relevant universe.” (/d.). Dr. Anderson chose to use two out of the seven webpages to construct his survey. He chose one webpage (“Ad Webpage”) “because it had the greatest number of views among the SignNow webpages” at issue and another webpage as an example of one of the several “form webpages” (“Form Webpage”). (D.I. 161-3 at 9 of 320 { 22). Even if SignNow offers “document storage services,” the Ad Webpage heavily focuses on SignNow’s e-signature capabilities. All four of the colored tiles at the top of the page reference signing or signatures, as does the customer testimonial quotation further down the page. (/d. at 64, 65 of 320). Other places on the Ad Webpage read, for example, “Increase your productivity by switching to a secure and reliable eSignature solution,” and “signNow is an award-winning eSignature
solution.” (Ud. at 65, 66 of 320). The top of the Form Webpage references “document generation” and “Production,” but the text otherwise also focuses on signing capability. (Jd. at 70, 73-74 of 320). At least for the intended audience of the two webpages used as the test pages, it seems that “document storage” is auxiliary to the e-signature capabilities. The extent to which SignNow’s potential document-storage-but-not-eSignature customers were excluded from the survey pool does not make the universe surveyed so underinclusive that the survey is inadmissible. Likewise, airSlate has not shown that its pool of customers includes a significant number of “individuals” who are not purchasing SignNow services as part of their job. The subscription packages available indicate that the most likely customer for SignNow’s services would be a business customer. An overbroad universe can be as much of a problem as an underinclusive one. When a survey includes respondents who are “not potential consumers,” those respondents “may well be less likely to be aware of and to make relevant distinctions when reading ads than those who are potential consumers” and would therefore fail to “make ‘crucial’ distinctions in the likelihood of confusion testing.” Smith v. Wal-Mart Stores, Inc., 537 F. Supp. 2d 1302, 1326 (N.D. Ga. 2008) (citation omitted). Likely purchasers of a product are more likely to be discerning and less likely to be confused. Given the evidence available, the customers most likely to purchase a SignNow subscription would arguably be the ones looking for eSignature services as part of their job. airSlate can cross-examine Dr. Anderson about his choice to exclude portions of the population and present evidence about its customer base to counter this assumption.
2. Biased survey design airSlate argues that the survey exhibits design bias because Dr. Anderson knew about the instant litigation and knew which results would be favorable to Inkit, leading to his inclusion of Question 5.2 (D.I. 160 at 15). airSlate also argues that the results to Question 3 indicate “acquiescence bias” because 98.3% of respondents in the test group answered “Yes, the webpage does communicate or imply something about DocuSign and Inkit” in response to the question ‘Does or doesn’t the webpage communicate or imply anything about DocuSign and inkit?” (/d. at 16; see D.I. 161-3 at 14, 26 of 320).? Inkit notes that a “double-blind design,” like Dr. Anderson’s study, is part of “best practices.” Inkit argues, “There is no rule that the designer of the survey must have ‘no knowledge of the litigation or the purpose for which the survey was conducted,’ only that the interviewer have no knowledge of the litigation.” (D.I. 167 at 15 (citing Citizen Fin. Grp., Inc. v. Citizens Nat. Bank of Evans City, 2003 WL 24010950, at *3 (W.D. Pa. Apr. 23, 2003), aff'd, 383 F.3d 110 (3d Cir. 2004))). Inkit contends this requirement was satisfied because the interviewer, Prodege, did not know the purpose of the survey.!° (/d. at 16).
§ Question 5 was a multiple-choice question allowing one response, reading: “Which, if any, of the following statements does the webpage communicate or imply about DocuSign and inkit?” The potential responses were (1) “DocuSign and inkit are the same company”; (2) “DocuSign and inkit are not the same company, but are connected or affiliated with each other”; (3) “DocuSign and inkit are not the same company, and are not connected or affiliated with each other”; (4) “DocuSign and inkit have some other business relationship not listed above”; (5) “The webpage does not communicate or imply any of these statements about DocuSign and inkit”; and (6) “I don’t know.” (D.I. 161-3 at 14 of 320). ? 89.1% of the control group also answered “Yes” to Question 3. (D.I. 161-3 at 26 of 320). 10 Tn fact, the survey was “self-administered,” which meant that “respondents provided their own answers to the questions, and no interviewer was involved” during administration. (D.I. 161-3 at 17 of 320 { 38). 13
A “survey must be conducted independently of the attorneys involved in the litigation” and “[t}he interviewers or sample designers should . . . be trained, and ideally should be unaware of the purposes of the survey or the litigation.” Pittsburgh Press Club v. United States, 579 F.2d 751, 758 (3d Cir. 1978).!! airSlate does not, however, fault Dr. Anderson for being an interviewer or sample designer. airSlate faults him for being the survey designer. J agree with Inkit that the survey is not inadmissible because the survey designer was aware of the purposes of the litigation and had discussions with Inkit’s attorneys. I fail to see how a survey that is targeted to the appropriate population or subject matter could possibly be designed by someone with no knowledge of the litigation or the issue at hand. And even a survey designer who knows about the litigation and purpose of the survey can create an instrument that produces valid measurements when the survey is undertaken with appropriate mechanisms to minimize bias, particularly at the point of collection. In comparison, when an interviewer knows the purpose of the litigation or which party sponsored the study, this may inject bias into the study because an interviewer might subconsciously influence a participant as to what the “correct” answer might be. See Smith, 537 | F. Supp. 2d at 1330-31 (“[D]ouble-blind conditions are essential because if the respondents know what the interviewer wants, they may try to please the interviewer by giving the desired answer, and if the interviewer knows what his employer wants, he may consciously or unconsciously bias the survey through variations in the wording or tone of his questions.”); FED.
'1 Admittedly, Dr. Anderson designed the sample webpages for the survey, so he is the “sample designer” as well as the survey designer. However, the samples used are largely identical to two of the actual webpages at issue in this case. (see D.I. 161-3 at 8-9 of 320 22). The only difference present in the two control webpages compared with the two test webpages was the replacement of “inkit (DocuSign)” with “inkit” alone. (Ud. at 11 of 320 9 27). Further, airSlate does not raise any issues with the images used, only the survey design itself. 14
JUD. CTR., REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 409-11 (3d ed. 2011). But there was no interviewer involved in the administration of Dr. Anderson’s study to exert influence. The study participants opted to be a part of Prodege’s database and to take surveys generally. They then answered the questions of this survey on their own. Any “designer bias” present would simply be due to a fault of the survey instrument itself or Dr. Anderson’s later interpretation of the data.!? 3. Over-reliance upon leading questions The design of Dr. Anderson’s survey begins with open-ended questions, asking respondents (1) “What main messages does the webpage you just viewed communicate or imply?”; and (2) “What other messages, if any, does the webpage communicate or imply?” The combined responses to these two questions indicated that 1.3% of respondents (that is, 3 of 238 respondents, excluding double counting) viewing the two test pages featuring “inkit (DocuSign)” thought Inkit and DocuSign were the same company or affiliated companies. No respondents in either control condition (that is, 0 of 258 respondents) answered in this way. (D.I. 161-3 at 25 of 320; id. at 154 of 320). Question 3 asked, “Does or doesn’t the webpage communicate or imply anything about DocuSign and Inkit?” Of the 238 participants in the test conditions, 98.3% or 234 responded “Yes, the webpage does communicate or imply something about DocuSign and Inkit.” 89.1% or
12 Dr. Anderson did himself “classif[y] responses into themes and assig[n] a ‘code’ for each theme” to score the open-ended questions. (D.I. 161-3 at 24 of 320 n.47). However, as the responses to the open-ended questions showed little evidence in favor of Inkit’s claims, airSlate does not argue designer bias there. 15
230 of the 258 participants in the control conditions also responded affirmatively.!? (Jd. at 26 of 320; id. at 155 of 320). Only the respondents who answered “Yes” to Question 3 were presented with Questions 4 and 5. Question 4 was an open-ended question asking, ““What does the webpage communicate or imply about DocuSign and Inkit?” This time, 2.9% of the test condition respondents (that is, 7 of 238 respondents) and .8% of the control condition respondents (that is, 2 of 258 respondents) indicated that Inkit and DocuSign are the same company. (/d. at 27 of 320; id. at 156 of 320). Question 5 then asked, “Which, if any, of the following statements does the webpage communicate or imply about DocuSign and Inkit?” Question 5 was a multiple-choice question for which respondents were required to pick one of six possible responses. Of test condition respondents, 42.9%, or 102 out of 238, answered that “DocuSign and Inkit are the same company,” and 16.8%, or 40 out of 238, answered that they are not the same but are “connected or affiliated.” Of control condition respondents, 12.0%, or 31 out of 258, answered that “DocuSign and Inkit are the same company,” and 11.6%, or 30 out of 258, answered that they “are not the same company, but are connected or affiliated.” (/d. at 28 of 320; id. at 157 of 320). In Dr. Anderson’s analysis, he subtracted the percentage of control respondents who answered that Inkit and DocuSign were the same or affiliated from the percentage of test
13 airSlate contends that the affirmative responses from 89.1% of respondents in the control group render the survey invalid because “the control image did not display the disputed statement at issue, ‘inkit (DocuSign).’” (D.I. 160 at 6-7). Though the headers of the control images did not display “inkit (DocuSign),” there were references to both Inkit and DocuSign in the control image webpages. For example, the control image for the Ad Webpage said “sionNow is a great alternative to inkit” at the top and further down contains what appears to be a customer testimonial saying, “I was closing another deal and had an outstanding quote already sent through DocuSign ... .” (D.I. 161-3 at 67-68 of 320). A reasonable respondent could therefore respond that both the test and control images “communicate or imply something about DocuSign and Inkit.” 16
respondents who answered the same—a measure he calls the “Net consumer perception.” (id. at 30 of 320). Inkit argues that the survey was well-designed to properly filter out participants who got “no message” from the advertisement, only asking Question 5 “at the end of the funnel” to “get at the ultimate issue: whether the participants are confused or not by the advertisement.” (D.I. 167 at 13, 15). airSlate argues that the results to every question other than Question 5 “show that consumer perceptions of Defendants’ communications have virtually no impact.” (D.I. 160 at 8). airSlate argues that the open-ended questions (i.e. Question 4) are a “more accurate and reliable measure of consumer perceptions” because they “did not improperly cue respondents or lead them to any desired answer[.]” (/d. at 8). “A survey is not credible if it relies on leading questions which are ‘inherently suggestive and invite guessing by those who did not get any clear message at all.’” Johnson & Johnson- Merck Consumer Pharms. Co. v. Rhone-Poulenc Rorer Pharms., Inc., 19 F.3d 125, 134 Gd Cir. 1994) (citation omitted). “Suggestive questions render a survey unreliable by creating ‘demand effects’ or ‘cues’ from which a respondent can ‘infer the purpose of the survey and identify the ‘correct’ answers.’” Valador, Inc. v. HTC Corp., 242 F. Supp. 3d 448, 465 (E.D. Va. 2017). Demand effects “sugges[t] to respondents, at least implicitly, that they should believe there is at least some sort of relationship between the different items when the possibility might not have □
even occurred to the vast majority of consumers who see the items.” Simon Prop. Grp. L.P. v. mySimon, Inc., 104 F. Supp. 2d 1033, 1048 (S.D. Ind. 2000). I agree with airSlate that Question 5 is inappropriately suggestive and leading. As airSlate’s rebuttal expert, Mark Keegan, explained, “[Q]uestions that fail to provide an equal number of answer options favoring each side ‘can telegraph to the respondent the nature
of the response expected (perhaps even desired) by the investigator [and] respondents with no firm answer would be more inclined to select one of the responses indicating agreement.’” (D.I. 168-1 at 352 of 417 § 60 (citation omitted)). In Dr. Anderson’s survey, three out of six of the answer choices of Question 5 favored Inkit, while one favored airSlate and two were neutral. (id. at 352 of 417). Additionally, a well-designed consumer survey should be “properly ‘filtered’ to screen out those who got no message from the advertisement[.]” Novartis Consumer Health, Inc. v. Johnson & Johnson-Merck Consumer Pharms. Co., 290 F.3d 578, 591 (3d Cir. 2002). Question 3 was a filter question, only allowing participants who understood the webpage to “communicate or imply anything about DocuSign and Inkit” to continue on to Questions 4 and 5. When answering Question 4, only 2.9% of respondents in the test condition interpreted the webpage as communicating a message that Inkit and DocuSign were the same company or affiliated companies. (D.I. 161-3 at 27 of 320). When provided with answer choices that suggested Inkit and DocuSign might be connected, however, that number jumped to 59.7% in Question 5. (d. at 28 of 320). The survey was properly filtered up until Question 4. However, Question 5 was improperly leading and suggestive, creating demand effects that appear to have encouraged participants to find a connection between Inkit and DocuSign that they previously had not contemplated. Accordingly, Question 5 of Dr. Anderson’s survey is unreliable. On the question of fit, testimony on Questions 1 through 4 would be helpful to the trier of fact, but testimony on Question 5 would not. Schneider, 320 F.3d at 404. I find that Dr. Anderson may testify about the survey results of Questions 1 through 4 but may not testify about Question 5 or rely upon Question 5 for his ultimate conclusions about
consumer confusion. airSlate’s motion to exclude the testimony of Dr. Anderson and his survey is GRANTED in part and DENIED in part. C. Federal Rule of Evidence 403 airSlate argues both experts should be excluded under Federal Rule of Evidence 403. (D.I. 157 at 17-18; D.I. 160 at 20). Any Rule 403 objections can be raised at trial. IV. CONCLUSION An appropriate order will issue.