On February 25, 2026, the Federal Court of Appeal (per De Montigny C.J., Webb J.A. and Pamel J.A.) (“FCA”) upheld a Federal Court decision (discussed in our previous blog post) that Promotion in Motion Inc. (“PIM”)’s trademarks SWISSKISS and SWISSKISS & Design (in association with Swiss chocolate) were not registrable in light of Hershey Chocolate & Confectionary LLC (“Hershey”)’s registered KISS and KISSES marks.
This decision highlights the distinction between certification marks and traditional trademarks. It is also a good example of the Court’s critical approach to survey evidence in trademark cases. The FCA ultimately agreed that the online surveys in this case were flawed and inadmissible. Nevertheless, the FCA acknowledged that online surveys, while not perfect, can be useful when properly designed and implemented and left the door open for online survey evidence in future trademark confusion cases.
Promotion in Motion, Inc. (PIM Brands, Inc.) v. Hershey Chocolate & Confectionery LLC, 2026 FCA 40
The two primary issues on appeal were:
- Did the Federal Court err in focusing on the KISS element of the marks and in holding that the SWISS element is a descriptive certification mark that cannot be distinctive?
- Did the Federal Court err in concluding that the survey evidence filed by the parties is inadmissible?
Issue #1: Descriptive Certification Marks
In the underlying opposition proceeding, the Trademarks Opposition Board (the “Board”) considered the SWISS component of the SWISSKISS marks descriptive. It therefore focused the degree of resemblance factor in the confusion analysis on the KISS element of the SWISSKISS marks. The Board ultimately found a likelihood of confusion between PIM’s SWISSKISS marks and Hershey’s KISS marks.
To counter the Board’s finding that SWISS was descriptive, PIM filed evidence of registered Canadian certification marks for SWISS, SUISSE and SWITZERLAND in association with chocolate products.
PIM argued that these certification marks were evidence of the distinctive character of the term SWISS. Since the SWISS certification marks are “trademarks” under the Trademarks Act (“Act”), PIM argued that the SWISS certification mark within the SWISSKISS marks is itself a specific type of trademark, and not just a geographical descriptive term. PIM claimed that mischaracterizing SWISS as descriptive skewed the confusion analysis by wrongly putting the emphasis on the KISS component of the SWISSKISS marks.
The FCA rejected PIM’s arguments, finding no error in the Federal Court’s decision. The FCA confirmed that certification marks registered under s. 25 of the Act, like the SWISS certification mark, are different in nature from traditional trademarks. Geographical certification marks, like SWISS, identify goods that are produced in certain places and not ordinarily registrable as trademarks under section 12 of the Act which precludes registration of marks that are clearly descriptive of the place of origin of the associated goods.
Despite their descriptive nature, descriptive certification marks are “saved” by section 25 of the Act which specifically allows for the registration of a descriptive certification mark so long as certain other conditions are met.
For the FCA, the existence of the registered certification mark for SWISS did not alter the descriptive character of the term. To the contrary, it was clear that “the certification marks SWISS, SUISSE, and SWITZERLAND are descriptive of the place of origin of wares from Switzerland…and that these certification marks were registrable only because of the presence of section 25 in the Act and not because they have any distinctive character.”[1]
It was therefore not an error to characterize SWISS as descriptive or focus on the KISS part of the SWISSKISS marks in the confusion analysis.
Issue #2: Online Survey Evidence
In the decision below, the Federal Court found several validity and reliability issues with the design of PIM’s surveys. These flaws led the Court to conclude that the survey evidence, submitted for the first time before the Federal Court, was inadmissible.
The FCA affirmed two of the Federal Court’s three criticisms of the PIM surveys.
First, the FCA found no error with the Federal Court’s criticism that it was inappropriate that the questions repeatedly referenced “Swiss chocolates” as opposed to simply “chocolates”. The Federal Court had held that repeated use of “Swiss chocolates” created a “priming bias” that reduced associations with Hershey.
PIM argued that “Swiss chocolate” was an appropriate descriptor in the survey questions since that was the category of product at issue. The FCA disagreed as PIM had not demonstrated that Swiss chocolate was an entirely different product category than chocolate from other countries. To the contrary, there was evidence that chocolate retailers in Canada sold Swiss chocolates close to chocolates of other origins.
Second, the FCA found no error in the Federal Court’s criticism of the ability of survey participants to return to earlier screens in the surveys. A survey must simulate the imperfect recollection of a trademark of a casual consumer in a hurry. The FCA held that a valid survey will not allow participants to go back and see an image once it has been removed from their view. The Federal Court therefore did not err in excluding these surveys since they allowed a user to see a prior screen either via a “back button” embedded in the surveys themselves or because they did not prevent the use of the “back button” built into standard internet browsers.
On the third criticism, the FCA departed from the Federal Court’s decision. The Federal Court had excluded the surveys because they were conducted online with no “person in the loop” who could verify that the survey participant was who they claimed to be or to ensure that the participant was free from external influence.
The FCA recognized these concerns inherent in online surveys but disagreed that such surveys are therefore fundamentally flawed and inadmissible unless there is a “person in the loop” or recording of the participant filling out the survey.
The FCA noted that phone surveys are open to the same criticism, yet are commonly used in courts today. It also pointed to evidence that online surveys, despite their flaws, offer advantages compared to in-person surveys such as cost-efficiency, wider coverage of populations, and avoidance of potential “interviewer effects”. The FCA also pointed to evidence of strategies that experts can implement to mitigate these concerns and improve the reliability and validity of online surveys.
The FCA agreed that this flaw is a concern inherent in online surveys but was not so critical as to be an issue of admissibility rather than weight. The FCA commented that internet surveys, though imperfect, can provide valid evidence to inform the confusion analysis when properly designed and implemented and that the admissibility of online survey evidence must be determined on a case-by-case basis. Although the FCA found that the Federal Court erred in excluding these surveys on this basis, this error was not determinative or overriding.
In the decision below, the Federal Court noted that had it admitted the survey evidence, that evidence would have supported Hershey’s case since the level of confusion revealed in the surveys was not insignificant and fell within the lower range of the rate of confusion Canadian Courts have found to be sufficient to establish a likelihood of confusion. The FCA found no error in this aspect of the Federal Court’s decision and found no basis for PIM’s assertion that a 10% threshold is typically considered necessary to show confusion.
Conclusion
The FCA found no palpable and overriding error in the Federal Court’s analysis of descriptive certification marks and in finding that the SWISS certification mark has no impact on the confusion analysis. The FCA also held that the Federal Court did not err in rejecting the survey evidence. The FCA dismissed the appeal with costs.
[1] Para 41.