In a trademark infringement litigation or proceeding before the Trademark Trial and Appeal Board (the Board), consumer surveys are often regarded as the most direct and persuasive evidence of the likelihood of confusion. While courts acknowledge that consumer surveys are expensive, some courts are presuming such evidence would be unfavorable in cases involving parties that could afford to conduct a survey, yet chose not to present such evidence. On the other hand, the Federal Circuit has made clear that surveys are not necessary in an administrative proceeding, however, “there may be cases in which the evidence of likelihood of confusion is not strong enough to support a finding to that effect in the absence of survey evidence.” For these reasons, litigants often use surveys to prove their case in both contexts.
How much confusion is enough? Professor McCarthy notes in McCarthy’s on Trademarks and Unfair Competition that, “While survey percentages demonstrating confusion levels over 50% are almost always viewed by courts as persuasive evidence of likely confusion, figures in the range of 25% to 50% have been relied upon as support for a finding of a likelihood of confusion” and that at least one court has observed that “even 11% of a national market of millions of consumers constitutes a very large number of confused consumers.”