Survey evidence can be used in trademark disputes to establish consumer perception and brand power. A recent Federal Court decision provides guidance on factors that may influence the admissibility of online survey evidence.

The Federal Court’s Decision

In Promotion In Motion Inc. v Hershey Chocolate & Confectionery LLC, 2024 FC 556, the applicant PIM Brands appealed the Trademarks Opposition Board's refusal to register trademark applications for SWISSKISS and SWISSKISS & Design (collectively, the SWISSKISS Trademarks) covering "Chocolate of Swiss origin". Hershey Chocolate & Confectionary LLC had successfully opposed registration of PIM's applications based on its own prior registrations for KISSES and KISS, covering "solid chocolate candy with or without ingredients such as nuts" and "chocolate candy".

The Federal Court ultimately held that two online surveys filed as evidence by PIM in the appeal failed to meet the criteria for validity and reliability necessary for this type of evidence to be admitted. Experts retained by PIM had administered surveys to measure consumer perception as to source of the SWISSKISS & Design logo. The surveys were conducted by way of an Internet-based questionnaire with participants who were Swiss chocolate purchasers. Test group participants were shown PIM’s SWISSKISS & Design logo, while control group participants were shown a modified version of the logo which replaced “KISS” with “WISH”, as illustrated below, drawn from the Court’s published decision:

Participants in both the test and control group answered the same series of questions, including questions testing their knowledge about the products and origin of source associated with the logo. Two surveys of this design independently concluded that participants did not associate Hershey as the source of the product branded with the SWISSKISS & Design logo, to any statistically relevant degree.

The Federal Court (Tsimberis J) held that these online surveys failed to meet the criteria of (1) validity (whether the right questions were put to the right pool of survey participants in the right way and in the right circumstances to produce the evidence sought) and (2) reliability (whether the survey results were reproduceable).

Validity

The Court found that the survey lacked validity because:

• the survey questions were flawed given that they repeatedly included mention of “Swiss chocolate”, which created a priming bias in the results;
• the manner in which the questions were asked was flawed. The Internet-based online survey format did not permit interviewers to attest that the intended participant answered the questions and did so without consulting any other devices (like a cell phone). This contrasts with in-person surveys, where participants are typically monitored by the interviewer; and
• the survey’s design was flawed because there was a risk that participants could use the “back button” in their Internet browsers. Clicking the “back button” would enable participants to repeatedly view the mark at issue, compromising the nature of the evaluation (which seeks to assess a consumer in a hurry, with an imperfect recollection of the impugned mark).

Reliability

The Court concluded that these validity concerns also undermined the ability for the same results to be reproduced. It was also unclear whether the online survey could be statistically generalized to the broad population of Swiss chocolate consumers because it consisted only of those who volunteered to be part of an online panel.

The Court ultimately concluded that the survey evidence was inadmissible.

Key Takeaways

The Federal Court recently expressed in Energizer Brands, LLC v Gillette Company (2023 FC 804) that survey evidence can be useful. The practical difficulties of implementing in-person surveys in the modern digital world, paired with the pitfalls of online survey evidence described in the PIM decision, may leave parties with a difficult choice about whether or not to rely on survey evidence. A party intending to rely on survey evidence should exercise caution when designing and conducting online surveys, and should consider means to limit potential vulnerabilities to validity and reliability, for example, by recording a video of participants completing the survey, as suggested by the Court.