Missing actual marketplace realities in confusion surveys can be fatal

United States of America
In THOIP v The Walt Disney Company (February 9 2010), the District Court for the Southern District of New York has excluded the confusion survey of plaintiff THOIP on the grounds that it did not recreate marketplace realities. 
In this case, THOIP claimed rights in an unregistered trademark consisting of the words 'Little Miss' with design elements in “big, bold capital letters”, plus the depiction of a character. THOIP's mark has allegedly been widely used in books, television series and videos, along with related merchandise. Merchandise sold using THOIP's mark includes a line of t-shirts sold in boutiques and prominent clothing retail chains in the United States. 
THOIP sued The Walt Disney Company under Section 43(a) of the Lanham Act, 15 USC §1115(a), alleging that a t-shirt line marketed as 'Little Miss Disney' and using the 'Little Miss Bossy', 'Little Miss Perfect', 'Little Miss Sassy' or 'Little Miss Wicked' wording, along with depictions of Disney characters, violated THOIP’s unregistered trademark rights. The Little Miss Disney shirts were sold at Disney theme parks, at the Disney store and at certain retailers - although allegedly in Disney departments. Both the THOIP shirts and the Disney shirts were also purportedly sold online by the same e-commerce outlets. In response to THOIP’s claims, Disney denied that its shirts created any marketplace confusion. To support their respective contentions, both parties undertook likelihood of confusion surveys. 
THOIP offered a survey by retained expert Dr Gary Ford. The survey used a “sequential array” survey in which respondents were shown a THOIP shirt and then taken to a separate room where they were shown five shirts, including an allegedly infringing Disney shirt, or alternatively a control shirt, and four non-infringing shirts. Based on the sequential survey, Ford found that consumers perceived the Disney shirts to be from the same source as the THOIP shirts.
Disney submitted the declaration of Dr Itamar Simonson to critique the Ford survey, as well as a consumer survey by Dr Myron Helfgott, demonstrating no likelihood of consumer confusion based on an Eveready survey. In the Disney survey, survey respondents were shown an allegedly infringing Disney shirt or, alternatively, a control shirt, and then queried about source, association and permission. This survey concluded that there was “virtually no evidence of confusion”. Thus, the survey results were diametrically opposed to those of THOIP’s survey expert.
On a motion to exclude THOIP's survey, the court found that this survey did not emulate marketplace conditions and, thus, should be excluded. The court recognized that the touchstone for evaluating the admissibility of a survey is whether the survey “simulated the actual marketplace conditions in which consumers encountered the parties’ products so as to be a reliable indicator of consumer confusion”. Moreover, the court acknowledged that a survey, by its nature, is an “imperfect mirror of actual customer behaviour” and that, normally, errors in survey methodology go to weight of the evidence. However, it held that a survey should be excluded when “it is likely to be insufficiently probative, unfairly prejudicial, misleading, confusing or a waste of time". 
After acknowledging that the parties’ products were directed at the same set of consumers, the court focused on how the surveys replicated actual marketplace conditions. The court found that THOIP’s sequential presentation survey reflected a real-world situation were both marks were likely to be evaluated by consumers side by side. When the products are the same, but sold primarily in different stores, the court held that the Eveready approach, and not a sequential survey, is the appropriate design. The court held that THOIP had not presented sufficient evidence to support its survey approach that consumers encountered the products in the actual marketplace side by side. Moreover, the court noted that the products coupled in THOIP's survey were chosen because they resembled each other, not because consumers encountered these products next to each other in the actual marketplace. 
Further, the court held that THOIP's survey did not use an adequate control. Therefore, the survey did not provide any type of benchmark. The control in THOIP's survey consisted of t-shirts with a Disney character, but without any words. The court found this difference critical and, as a result, held that the control also failed to replicate actual marketplace conditions.
In any trademark infringement case relying on consumer confusion surveys, the survey should be justified by marketplace realities. This case highlights the importance of understanding the realities of the marketplace before adopting a survey design. 
Rochelle D Alpert, Morgan Lewis & Bockius LLP, San Francisco

Unlock unlimited access to all WTR content