court declines to exclude false ad survey despite variance from TM survey practice

Simpson Strong-Tie Co. v. MiTek Inc., 2023 WL 137478, No. 20-cv-06957-VKD (N.D. Cal. Jan. 9, 2023)

Simpson sued MiTek for Lanham Act and state law false advertising/passing off, and for copyright infringement. The court granted in part and denied in part cross-motions to exclude survey-related expert evidence.

Simpson makes structural connectors for use in building construction, and assigns alphanumeric product names for each of its products. Each product name has a “part name” (a letter or combination of letters designating the product line) and a “model number” (additional numbers and letters that distinguish parts with different attributes).

MiTek sells structural connectors that compete with Simpson’s. MiTek allegedly uses product names that are identical or similar to Simpson product names and also uses Simpson’s product names as “reference numbers” for MiTek’s own products. This allegedly deceives consumers into believing that the companies’ products are equivalent or interchangeable when they are not, or that MiTek’s products are actually Simpson’s products.

Simpson’s survey expert Wallace conducted four surveys that examined, inter alia, whether MiTek’s uses led consumers to believe that the products were equivalent and/or have the same attributes; that the sources of these products were the same or affiliated with one another; that Simpson endorses MiTek’s reference to Simpson’s product; and whether confusion regarding these issues affected purchasing decisions. MiTek’s expert Franklyn critiqued this evidence.

MiTek argued that Wallace failed to use an accepted survey methodology; Simpson responded that, in false advertising cases such as this, the Eveready and Squirt methodologies do not necessarily apply, and that the survey used must be tailored to the nature of the advertising claim at issue.

“Where there is no single, generally accepted methodology, the distinction between admissible and inadmissible survey evidence is difficult to make.” But the Ninth Circuit “has repeatedly cautioned district courts against excluding (rather than appropriately weighing) even seriously flawed survey evidence.”

MiTek challenged (1) the universe as overinclusive of people in the construction industry and underinclusive of do-it-yourselfers, (2) the sample (these two criticisms went only to weight), (3) the questions, (4) the reporting of the data, and (5) the statistical analysis of the data.

Questions: In each survey, respondents first got information about the stimulus, e.g.:

“MiTek sells a structural connector called the ADTT-TZ. You are about to view a product label for MiTek’s ADTT-TZ product. Please base your answers to the following questions on the information on the label itself. Please view this label as if you were considering purchasing or specifying this product for your current work.”

They were then asked questions such as:

• Based on this product label above, do you believe that MiTek’s ADTT-TZ product is equivalent to the DTT1Z product?

• Based on this label, do you believe that MiTek’s ADTT-TZ product has the same attributes, such as load capacity, number of required fasteners, and existence of code reports, as the DTT1Z product?

• Based on this label, do you believe that MiTek sells both the ADTT-TZ product and the DTT1Z product?

• Based on this label, do you believe that the company that sells the DTT1Z product endorses or approves of MiTek’s reference to the DTT1Z product?

Additional questions asked how respondents would behave if they learned certain additional information, such as:

• If you learned that the company that sells the DTT1Z does not endorse or approve of MiTek’s reference to the DTT1Z product, would that cause you to specify or purchase the DTT1Z product instead of MiTek’s ADTT-TZ?

• If you learned that the DTT1Z product is not sold by MiTek, would that cause you to specify or purchase the DTT1Z product instead of MiTek’s ADTT-TZ?

• If you learned that MiTek’s ADTT-TZ product has a load capacity of 820 lbs. for Douglas Fir/Southern Pine, while the DTT1Z product has a load capacity of 910 lbs., would that cause you to specify or purchase the DTT1Z product instead of MiTek’s ADTT-TZ product?

• If you learned that MiTek’s ADTT-TZ product requires 10 fasteners to achieve its load capacity while the DTT1Z requires 6 fasteners to achieve its load capacity, would that cause you to specify or purchase the DTT1Z product instead of MiTek’s ADTT-TZ product?

The answers were always presented in the same order; the “Yes” response was always favorable to Simpson and unfavorable to MiTek.

MiTek argued that the use of closed-ended questions instead of an open-ended question asking respondents what “Ref #: DTT1Z” means on the MiTek product label suggested both that DTT1Z was a reference to another company’s product and an equivalence claim. And, where close-ended questions are used, invariant response order is particularly troubling. Although this was a serious problem, “given the flexibility that survey professionals apparently believe is necessary in constructing a survey instrument intended to test for false or misleading advertising, the Court cannot say definitively that Mr. Wallace’s use of these types of questions renders his surveys so unreliable as to permit or require exclusion under Ninth Circuit law.” The flaws would be weighed at the forthcoming bench trial.

Secondary meaning: In each survey, questions about the stimulus were followed directly by questions regarding whether and to what extent respondents associate certain product names with Simpson or MiTek as a source of the product. Respondents were asked if they recognized certain Simpson product names (in random order), and if so, they were given space to write in which company or companies they associated with each product name. MiTek argued that, because the questions immediately followed the sequence of questions designed to test for advertising confusion in which respondents repeatedly were shown stimuli including MiTek’s brand name, the surveys improperly conditioned respondents to mention MiTek.

Simpson argued that there was no authority requiring a separate survey, and the survey results themselves suggest not much leadingness,  as only 21 respondents out of 400 identified MiTek in response to the question regarding the products’ source. But these 21 respondents represent over 20% of the respondents who indicated any awareness of the product names at all, and only 53 respondents recognized Simpson as the source of its own products. Although MiTek’s arguments were persuasive, they again went only to weight.

What about controls? For false advertising, the control group respondents were exposed to the same four MiTek stimuli but with the reference numbers removed. The results were similar to those of the test group. For secondary meaning, the control group respondents were exposed to six fictitious product names instead of actual Simpson product names and were asked the same questions.. Here too, the results were similar. Simpson’s expert argued that this could be “the result of MiTek’s longstanding deceptive marketing practices that have created pre-conceived notions throughout the industry that the MiTek products and Reference Number (Simpson) products are equivalent.” He suggested that “these control respondents truly believe MiTek’s deceptive messaging and ... this deception is so widespread that it persists even when the Simpson product names are removed from MiTek’s communications.” This testimony was unsupported and speculative and would be excluded. The expert “may not speculate as to the reasons for either set of survey results, as his report discloses no basis for the explanations he provides.”

MiTek’s rebuttal expert was not excluded merely because he conducted no survey of his own; that’s fine. Simpson also argued that his methodological concerns apply only to trademark infringement claims and not false advertising claims, and for this reason his opinions were unreliable. The court disagreed. However, because he didn’t have relevant expertise in the construction industry, certain aspects of his proposed testimony would be excluded or would be admitted contingent upon the record at trial, specifically testimony about the commonality of overlapping product names in the industry, about why MiTek chose particular naming or marketing strategies, or about how the relevant consuming public would react to alternative product names.