We've previously discussed some of the challenges of obtaining quality consumer survey evidence in trademark litigation in past posts, such as here and here. The persuasiveness and wisdom of a consumer survey was once again called into question, this time by the Federal Circuit in its October 30 opinion in Converse's long-running trademark litigation against Skechers, New Balance, and others. The appeals court identified a "series of errors" requiring remand, including that the International Trade Commission applied incorrect standards to both invalidity and infringement and erred in its secondary meaning analysis.

The case, which has been closely watched by trademark practitioners, started when Conversed filed a complaint in 2014 in the ITC alleging infringement of Converse's trade dress in the midsole of its All Star shoes: In June 2016, the ITC issued its final determination ruling in part that Converse's trademark was invalid based on its finding that Converse had not established that the mark had acquired secondary meaning, i.e. that consumers associate the sneaker design with Converse. The Federal Circuit reversed in part because the ITC never made a specific determination of secondary meaning as of the relevant date: Converse, as the party asserting trade dress infringement, needed to establish that its mark had acquired secondary meaning before the first infringing use by each accused infringer.

That brings us to the survey submitted by the ITC respondents to show that the mark had not acquired secondary meaning. The Federal Circuit began by noting the inherent problem in relying on a survey as evidence of historic secondary meaning—a contemporaneous survey for litigation may not accurately assess consumer attitudes at the relevant "point in the past." Here, the appeals court concluded that the survey was entitled to "little probative weight" to the issue of secondary meaning in the infringement analysis since the relevant point in time was 5-10 years prior to the survey (the survey may, however, have relevance to secondary meaning at the time of registration two years earlier).

The Federal Circuit also noted that, while there was no error in concluding that the survey does not establish secondary meaning, it was "unclear" why the ITC concluded that the survey supported the opposite, i.e. a lack of secondary meaning. The appeals court therefore provided the ITC the following guidance on remand: "unless the survey shows a lack of secondary meaning, there is simply a lack of survey evidence of secondary meaning—which is a neutral factor favoring neither party."

The dissent deemed the majority's discussion of the survey "pure dicta" but noted that if it were to consider the survey, it would be even less inclined than the majority to give the survey weight.

The Federal Circuit's opinion once again highlights the challenges in obtaining a relevant survey. In fact, between the various parties, the ITC ALJ had rejected eight other surveys because they used improper control and test images. Considering the expense of conducting a survey, trademark litigants should work closely with their survey expert to carefully craft a well-designed survey. Prudent litigants also would be smart to consider whether secondary meaning can be established through other non-survey evidence.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.