Survey Research Expert Witness’ Testimony Admitted Because the Court Found No Evidence of Actual Confusion among Respondents
Posted on November 28, 2024 by Expert Witness Profiler
In this trademark dispute, Defendants SuccessfulMatch.com and Successful Match Canada (both, Successful Match) sought to cancel on genericness grounds the registration of trademarks asserted by Plaintiffs Clover8 Investments and Reflex Media, Inc.
As Successful Match puts it, the trademarks in question “are used in connection with seeking a companion in the ‘Sugar Daddy’ and ‘Sugar Baby’ relationship space, online or through an App, which are designed for individuals seeking a specific type of arrangement typically between a young woman and a wealthy man who is often a millionaire.”
Each side retained an expert witness — Hal Poret for Successful Match and Brian Sowers for RMI — to conduct a consumer survey meant to determine whether the disputed marks are generic or not. Each side then attacked the other expert’s survey work as junk science that ought to be excluded from use in this case under Federal Rule of Evidence 702 and Daubert.
The Court convened a concurrent expert evidentiary proceeding to hear directly from Poret and Sowers about why they disagreed and why the other’s survey work might be so deficient as to warrant exclusion.
Poret and Sowers agreed that the top objections were whether (1) the scope of the “universe” of the relevant purchasing public each expert surveyed was appropriate; (2) it was appropriate to use examples tailored to the sugar-dating market in the survey’s “mini-test”; (3) the definitions for “brand” and “generic” used in the surveys were appropriate; and (4) the control terminology for “generic” names used in the surveys were appropriate.
Survey Research Expert Witnesses
Hal Poret is a public opinion researcher with a master’s degree in mathematics and a law degree from Harvard Law School. Since 2004, Poret has personally designed, supervised, and implemented over 350 consumer surveys concerning consumer perception, opinion, and behavior, including surveys relating to false advertising, claim substantiation, and trademark/trade dress matters. He has personally designed numerous studies that have been admitted as evidence in legal proceedings and has testified as an expert in over 50 proceedings before U.S. District Courts, the Trademark Trial and Appeal Board, and NAD.
Brian Sowers is a principal and testifying survey expert at Applied Marketing Science (AMS). He is responsible for the delivery of survey research and expert testimony in civil cases. Sowers also supports affiliated academic experts and economic expert witnesses in matters in which consumer opinions and behaviors are an important determinant of liability and damages.
Discussion by the Court
The parties do not disagree that the genericness inquiry looks to whether the consuming public “uses or understands the term sought to be protected” as a generic name for the “services set forth in the certificate of registration.”
Rather, the dispute is about how to define the consuming public and populate the relevant survey universe, namely whether the proper universe is comprised of consumers of online dating or matchmaking services generally, which would match the description of the services designated in RMI’s registrations, or just consumers of “luxury” dating or “sugar dating” services.
The parties do not agree about the nature of the services offered in connection with the marks. Successful Match says RMI “expanded the scope of [their] services to include a more general universe of consumers” by “mov[ing] away from the sugar daddy dating space . . . and shift[ing] the focus to connecting people in more typical relationships.” RMI says that they offer a “luxury dating service” and nothing suggests they offer “a mainstream or traditional dating site.”
The Surveys’ Mini-Tests
Poret and Sowers each conducted a “Teflon survey,” which is a type of consumer survey that is “essentially a mini-course in the generic versus trademark distinction, followed by a test” wherein the survey-taker is asked to categorize various names or phrases as a “brand” name or term or “generic” name or term. Teflon surveys typically include a “mini-test” before the main test to ensure that the respondent properly understands and can apply the “brand”/”generic” distinction.
Sowers’ Mini-Test Questions
Poret criticized Sowers’ questions in the mini-test, which asked respondents to categorize as “brand” or “generic” the terms “Networking Website” and “Sugardaddie.” Poret believes these questions biased the survey results because they would lead to the exclusion of respondents who categorized “Sugardaddie” as generic, and so artificially tilt the results toward those most likely to say that RMI’s asserted marks are a brand.
At the concurrent evidentiary proceeding, Sowers stated that he tailored to the case by requiring respondents to apply the “brand”/”generic” distinction to “Networking Services” and “Sugardaddie,” the latter of which being a mark he thought was “a pretty prominent player in the marketplace” after conducting research on the sugar-dating space.
The thrust of Poret’s critique is that the term Sowers chose to “tailor” his mini-test is “as debatable as what the survey is about in the first place.” The Court held that the shortfall of this criticism is that an objection to the specific words Sowers used in the mini-test goes to “follow-on issues of . . . survey design” and not the threshold question about whether the survey was “conducted according to accepted principles.”
Indeed, at the evidentiary proceeding Poret stated that “it’s not the mere fact of including something in the mini-test that relates to the category that is the fatal flaw. . . . It’s the specifics of [Sugardaddie].” The Ninth Circuit has characterized “leading” or “slanted” questions as “technical inadequacies” that “bear on the weight of the evidence, not its admissibility,” and so the Court concluded that this specific objection to the word Sowers used in the mini-test most closely resembles those sorts of challenges.
Poret’s Common Term And Test Examples
Sowers said the “fatal flaw” in Poret’s survey was that its wording caused respondents to misunderstand the “brand”/”generic” distinction
For his part, Sowers said the “fatal flaw” in Poret’s survey was that its wording caused respondents to misunderstand the “brand”/”generic” distinction, which is said to render the survey irrelevant. Sowers specifically took issue with the prompt at the start of the main test, which told respondents they would need to “answer whether you think that term is . . . a common term that identifies a type of dating, matchmaking, and social introduction mobile app or website.” In Sowers’ view, the definition of a “common term” as a “type” of a dating or matchmaking service would cause respondents to be confused about what a “common” term is as a concept.
Prior to the language to which Sowers objects, respondents were told that “common terms” are “terms that identify a type of product or service. Common terms primarily let the consumer know what the product or service is, not who makes it. These terms primarily identify a type of product or service.
Sowers acknowledged he had no problems with that initial definition, so his objection appears to be directed at the shift from “identify a type of product or service” to “identify a type of [word or phrase for the specific product or service at issue].”
The Court held that this is a minor variation of no apparent import, particularly because respondents were required to read a prior, fuller definition and had to correctly apply the earlier definition before going on to the main test.
Sowers also did not present any evidence of actual confusion among respondents
Poret maintained that several of the main test’s questions were designed to control for potential confusion and that the overall results demonstrated that respondents properly identified brands versus generics. Sowers suggested that the discrepancy between respondents who identified “relationship” as a common term (96%) and “luxury dating service” as a common term (76%) was evidence of confusion. But well more than 50% percent of respondents agreed that “luxury dating service” was common, and a 20% variation in responses is not significant when there is more than majority concurrence.
Sowers also said that the terms Poret used as “common” controls confused respondents about the proper meaning of “generic” or “common.” The Court did not find it persuasive.
Survey respondents were instructed that a common term “let’s the consumer know what the product or service is” and that it “identifies a type of dating, matchmaking, and social introduction mobile app or website.” For each question in the main test, the term in question was capitalized at the top, and right below it the question read: “In the context of dating, matchmaking, and social introduction mobile apps or websites, do you think this is a . . . [(a)] Brand term [(b)] Common term [(c)] Don’t know.”
Since Sowers did not show that the omission of the word “app” or “site” caused respondents to ignore or become confused about the prior definitions, the Court held that it is entirely speculative to say that a respondent considering the word “relationship” would not have understood that they were being asked whether “relationship” referred to a brand or type of dating app (e.g., an app for persons looking for relationships as colloquially understood versus other arrangements) or “singles” was a brand or type of social introduction app for single individuals.
Other Challenges
Each side offered a grab bag of other comments that also do not support exclusion. RMI objected to the survey’s use of the phrases “brand term” and “common term” instead of “brand name” and “common name.” This exalts form over substance. Although “term” and “name” may have somewhat different dictionary meanings, RMI adduced no evidence that respondents using everyday English would have appreciated that subtle difference. In addition, a plethora of federal-court decisions use “term” in the context of genericness, which demonstrates that the two words are functionally interchangeable for present purposes.
Poret’s reasons for selecting his common controls do not demonstrate the survey’s irrelevancy, as RMI suggests. As discussed, the Court held that there is no shortfall that bears on admissibility in Poret’s survey’s definitions, the failure to include trailing words like “site” or “app” do not provide a basis for excluding the survey, and so RMI’s reliance on one or two isolated statements in Poret’s report proves too much in the contexts of the report and the concurrent proceeding.
Lastly, Successful Match’s suggestion that Sowers’ survey is irrelevant because it tests secondary meaning rather than genericness does not carry the day. The Court held that Sowers’ focus on consumers of “sugar dating” services does not warrant exclusion on this record, and the qualification rate alone does not establish irrelevancy. Because Successful Match does not contend that Sowers’s mini-test was actually testing secondary meaning rather than genericness and so should not have been included for that reason, the Court need not address the argument.
Held
The Court denied RMI’s motion to exclude the opinions of Hal Poret and Successful Match’s motion to exclude the opinions of Brian Sowers.
Key Takeaway:
The parties do not agree about the nature of the services offered in connection with the marks. Successful Match says RMI “expanded the scope of [their] services to include a more general universe of consumers” by “mov[ing] away from the sugar daddy dating space . . . and shift[ing] the focus to connecting people in more typical relationships.” RMI says that they offer a “luxury dating service” and nothing suggests they offer “a mainstream or traditional dating site.” In light of this dispute, the Court cannot conclude that either survey’s universe is comprised of respondents whose views would be irrelevant to the bottom-line question of whether the disputed mark, “taken as a whole, signifies to consumers the class of online [matchmaking or dating] services.
As a result, the Court cannot conclude that either expert, in choosing his respective survey universe, failed to reliably apply well-accepted principles in the field.
Case Details:
| Case Caption: | Reflex Media, Inc. Et Al V. Successfulmatch.Com Et Al |
| Docket Number: | 3:20cv6393 |
| Court: | United States District Court, California Northern |
| Order Date: | November 26, 2024 |
