top of page
  • johnmckeownblog

Not a Good Day for the Use of Survey Evidence in Trademark Cases

The Facts

Promotion in Motion, Inc. (PIM), PIM sought to register the trademarks SWISSKISS and SWISSKISS & Design depicted below for the goods “Chocolate of Swiss origin” (collectively, “SWISSKISS Trademarks”). Promotion In Motion Inc. v. Hershey Chocolate & Confectionery LLC 2024 FC 556

Hershey Chocolate & Confectionery LLC (Hershey) opposed PIM’s trademark applications based on several registered KISS and KISSES trademarks.  The opposition proceeded to a hearing. The Trademark Opposition Board decided the opposition based primarily on Hershey’s registration Nos. TMA733,263 for the KISSES trademark and TMA833,060 for the KISS trademark (collectively “Hershey’s Trademarks”), registered in association with the goods “solid chocolate candy with or without ingredients such as nuts” and “chocolate candy” respectively.

The Trademark Opposition Board

The Board held that PIM had not satisfied its burden on a balance of probabilities to show no reasonable likelihood of confusion between its SWISSKISS Trademarks and Hershey’s Trademarks.

The Hearing officer said the question was whether the average Canadian consumer, with an imperfect recollection of the trademark KISSES and/or KISS in association with chocolate, when they see a Swiss chocolate product bearing the trademark SWISSKISS (or SWISSKISS & Design), would think that the products come from the same source. If any scope of protection is given to the registrations of the Hershey’s Trademarks that question must be answered in the affirmative.

The SWISSKISS Trademarks consist of a descriptive term “SWISS” preceding the term “KISS” for which there are already registered rights in Canada in association with chocolate. It is reasonable to assume that consumers familiar with the trademark KISSES and/or KISS in association with chocolate would think that Swiss chocolate bearing the trademarks SWISSKISS and SWISSKISS & Design comes from the same entity as the owner of the registered marks, or has been, at the very least, approved, licensed or sponsored by the owner of those registered marks.

The Hearing Officer considered that PIM’s SWISSKISS Trademarks were not circumscribed by any particular colour or packaging style that might help to reduce the likelihood of confusion in the marketplace. To the contrary, the registration of the SWISSKISS Trademarks would entitle the PIM to depict it marks in any colour or get-up PIM might choose.

The Appeal

PIM argued that the Board committed a reviewable error in its finding that PIM had not satisfied its burden to show that there was no reasonable likelihood of confusion between its SWISSKISS Trademarks, as applied for, and Hershey’s registered KISS and KISSES Trademarks. In addition, it was asserted that the Board committed a reviewable error in its finding that PIM’s evidence was insufficient to determine the word “KISS” is generic for chocolate, and that the KISS and KISSES Trademarks were generic or otherwise devoid of distinctiveness in association with chocolate in Canada.

The Survey Evidence

PIM filed significant new evidence on the appeal. The first area of evidence related to a survey conducted online by way of an Internet-based questionnaire between September 1st and 24th, 2020, with four hundred and fifty-four (454) online Canadian participants who were Swiss chocolate purchasers who have bought from a retail store at least once in the past three (3) months or intend to buy in the next three (3) months. The participants were shown among other things the SWISSKISS & Design mark. The purpose of the research was to measure the source, if any, that consumers attribute to the SWISSKISS & Design mark, in particular, to measure the extent, if any, to which consumers would mistakenly infer that this mark is sourced by Hershey based on its use of the word “KISS” in the logo.

It was concluded that the majority of survey participants either did not identify the SWISSKISS & Design logo with a specific source or identified its source as a company with the same name as that shown in the logo. There was no statistically significant evidence of the mistaken inference that Hershey was the source of the product using the SWISSKISS & Design logo.

A second similar survey was carried out to assess the general impression associated  with the SWISSKISS (word) brand, and to measure to which extent, if any, purchasers of Swiss chocolate at a retail store mistakenly infer that SWISSKISS originates from Hershey. This survey concluded that:

1.There was no meaningful statistical evidence (2% spontaneous association and 1% association as the maker) that HERSHEY was associated with the SWISSKISS brand shown in the Test Group;

2.Plus, or minus the margin of error (1.77%) applicable to this proportion of 3%, this means that total mistaken inference with Hershey would range from 1.23% to 4.77%;

3.The association to Hershey for the Test Group was not statistically different than associations to other brands spontaneously mentioned such as Lindt or Nestle;

4.The awareness of Hershey and its Hershey’s Kisses chocolate product remained high (95% in the Test Group and 97% in the Control Group). The fact that only 3% of the Test Group associated SWISSKISS to Hershey further suggests consumers do not to any meaningful extent associate SWISSKISS to Hershey.

Hersey arranged to carry out a modified version of the surveys carried out on behalf of PIM This study concluded that:

  1. There was a 10% spontaneous association between Hershey and the SWISSKISS brand shown in the Test Group, well above unaided associations to other chocolate.

  2.  The trademarks KISS and KISSES benefit from a strong association to Hershey.

Hershey also attacked the PIM surveys on the basis the questionnaires created a priming basis by repeating “Swiss chocolate” rather than “chocolate” which lead to false or skewed conclusions. The Judge said that her familiarity with these types of trademark market research surveys used to help the decision-maker in trademark matters made her acutely aware of these issues.

The survey evidence put forth in both the Corbin Affidavit and Bourque Affidavit, like all expert evidence, must meet the following four conjunctive requirements set out by the Supreme Court of Canada for the Court to find it admissible:

  1. Relevance;

  2. Necessity in assisting the trier of fact (in the sense that the evidence is outside the experience and knowledge of a judge);

  3. The absence of any exclusionary rule; and,

  4. A properly qualified expert.

Relevance in survey evidence includes the requirements relating to validity and reliability. The survey must be valid (in that 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). The survey must also be reliable in that if it were repeated to a different sample in the pertinent population on a different occasion, it would produce the same results.

The Judge found that the right questions were not put to the PIM survey participants rendering the survey design and its results invalid. “Swiss chocolate” was mentioned to survey participants at least seven or eight times in the survey questions. “Swiss chocolate” was in the screening questions (three or four times depending on their answers). “Swiss chocolate” was again in each of the four key questions in the questionnaire, so potentially up to another four times as part of the questionnaire (depending on the answers and flow through the screener and questionnaire).

The systematic and repeated inclusion of “Swiss chocolate” in each of the survey questions would result in the participants being primed into thinking of brands of Swiss chocolates and providing responses that relate only to Swiss chocolate brands or companies.

The PIM surveys were also said to be flawed in that PIM’s experts could not assure the Court that the participants are the consumers in question and that the responses were given by the participants themselves (rather than another person (e.g., their child) filling out the survey for the registered panel member to obtain the honorarium); and, the participants were free of external influence and external information.

While PIM presented additional arguments, they did not involve errors of law but involved questions of mixed fact and law and in the Judge’s view there were no extricable questions of law in issue. The applicable standard of review on the balance of the alleged errors and their components was palpable and overriding error. PIM was not able to overcome this high threshold.


Many people accept that survey evidence should be used more frequently in trademarks cases as when properly presented it tends be objective. Ideally the method of approaching the problem should be more standardized and Internet surveys should be favoured as they tends to be cheaper. We need to find a practical way to deal with survey evidence.

If you have questions, please contact me at

Goldman Sloan Nash & Haber LLP 480 University Avenue, Suite 1600 Toronto, Ontario M5G 1V2 Direct Line: (416) 597-3371 Fax: (416) 597-3370 Email:

These comments are of a general nature and not intended to provide legal advice as individual situations will differ and should be discussed with a lawyer.

7 views0 comments


bottom of page