Peloton music library class action fails because consumers probably didn't see the claim

Passman v. Peloton Interactive, Inc., 2023 WL 3195941, No. 19-cv-11711 (LJL) (S.D.N.Y. May 2, 2023)

Interesting discussion of the way in which the objective reasonable consumer standard allows consumers classes to bring certain probabilistic claims, where some consumers might have different interpretations, although the court ultimately denies certification because of damages/price premium issues.

Plaintiffs’ claims under New York law were based on the offer of an “ever-growing” or “growing” library of live and on-demand studio classes, which offer was allegedly false because a bunch of classes were pulled from the library because of unlicensed music use. After it was sued in 2019, Peloton removed approximately 6,500 on-demand classes from its library, leaving approximately 7,000 classes available to its members.

The court initially held that problems with plaintiffs’ expert’s conjoint damages model went to weight rather than admissibility, since it was a model that was consistent with their theory of liability.

Materiality and falsity were common questions subject to an objective inquiry into how a reasonable consumer would react. Peloton argued that there was too much variation in how, when, and why consumers bought subscriptions. “Defendant’s argument confuses the question of whether a reasonable consumer would likely be misled by an allegedly false advertisement with the separate question—relevant where reliance is at issue—of whether an individual consumer was misled by the advertisement….  The inquiry demands an objective analysis of the understandings a reasonable consumer would draw from a challenged statement, not the significance of that challenged statement to an individual consumer’s purchasing decision.” While context is relevant to that, it’s ad context, not any possible context:

[W]hat is relevant to the analysis is not the number of products purchased, their exact identity, or the circumstances under which they were purchased; what matters is the nature of the allegedly false statement and how consumers interact with the false statement and therefore understand it. Thus, courts generally have no difficulty finding named plaintiffs typical of a class so long as the challenged statement is consistent across the class.

Peloton also argued that falsity wasn’t a common question because there wasn’t 100% agreement in plaintiffs’ survey on the meaning of “ever-growing”—the survey found that only 76.5% of respondents thought that “ever growing” meant “increase over time.” But misleadingness is an objective inquiry. “Evidence that actual consumers, in fact, interpreted the challenged statement in line with ‘the plaintiffs’ proffered theory of deception’ is relevant to, and may be necessary for, the ultimate conclusion that a reasonable consumer would have been deceived.” But  “the fact that consumers may have interpreted the statement differently does not preclude certification.” Differences in understanding might bear on injury, but falsity itself “cannot differ from case to case or be based upon whether the case is prosecuted on an individual or a class basis; it turns upon an objective analysis that applies across cases.”

Nonetheless, because causation, injury and damages weren’t common among the putative class members, individual questions predominated and a Rule 23(b)(3) class couldn’t be certified. The evidence submitted didn’t show a price premium or propose a methodology which could be used to demonstrate a price premium.

Peloton presented affirmative evidence that there was no price premium and “compelling evidence” that plaintiffs failed to carry their burden on predominance. The challenged claim wasn’t on a product label or package:

[N]o purchaser of a Peloton product need have been exposed to the Challenged Statement and the evidence suggests that many of the purchasers were not exposed to the Challenged Statement. The Challenged Statement did not appear in every Peloton advertisement and did not appear on all of Peloton’s marketing materials. Rather, the Challenged Statement appeared in a relatively small subset of Peloton’s advertisements, and did not appear in any television advertisements, which represented Defendant’s largest advertising channel during the Class Period. The Challenged Statement also appeared on only four of the 269 pages of Pelton’s website, the primary place where consumers purchase Peloton products.

Even when it did appear, the claim wasn’t alone or the most prominent. E.g.: “Experience unlimited access to the world’s best instructors anytime, anywhere, with 15+ daily live classes and an ever-growing library of 9,000+ classes available on-demand.”

Website tracking data suggested that, at most, 10.99% of website visitors could have been exposed to the statement, but that didn’t show that they actually saw/noticed it.Peloton’s expert did a survey about whether consumers even noticed the claim. She found that only 0.7%-1.3% of the test group noticed the statement and that there was no statistically significant difference between the test and control with respect to their conclusions about the Peloton products after viewing the webpages. And, despite its statements above, the court also gave weight to plaintiffs’ own survey that showed that people ascribed “different meanings”: while 76.5% understood that the number of classes in the Peloton library would increase over time, 61.8% of them expected the number of classes to increase because no classes would be removed from the library, while 31.4% believed that more classes would be added than removed. Also, 79.8% interpreted the statement to mean the number of classes would increase each day, week, or month, while 3.7% thought the increase would be annual (the remainder had no view). This wasn’t fatal: “materiality is an objective inquiry and thus subject to common proof.”

But this evidence does bear on whether the Challenged Statement could have caused a price impact. That many had differing views over what the Challenged Statement meant and took different meaning from it suggests both that the statement did not have a powerful marketing impact and that those who saw the Challenged Statement may not have interpreted it in such a way as to give rise to a price premium.

In addition, the price of each relevant Peloton offering remained constant both before, during, and for almost eighteen months after the class period. While those results “could perhaps be explained away if there were other confounding factors, including if Defendant offered something additional of value to consumers after the takedown or if sales fell markedly,” Peloton’s expert looked for other confounding factors and found none. The burden of demonstrating that there was a price premium (and thus that the predominance requirements of Rule 23(b)(3) have been met) was plaintiffs’ and they did not satisfy it. Plaintiffs’ conjoint model assumed that individuals saw and noticed the challenged statement to generate its price premium calculations, but that assumption wasn’t supported by the evidence. Plaintiffs’ survey also didn’t distinguish the value of Peloton’s library of classes—a Peloton innovation—from the value of its purported “ever-growing” size.

In addition, plaintiffs didn’t show that there was a model capable of measuring the damages attributable to their theory of liability, since their model conflated the existence of a library with its ever-growing size. “Courts routinely reject price premium methodologies under Comcast when the proposed methodologies do not attempt to isolate the premium due only to the allegedly misleading marketing statement.” The model also didn’t consider the impact of supply-side factors. “Because Defendant might have responded to the decreased demand by producing fewer Peloton products, the damages of the putative class members would presumably be less.” Nor did the model “consider how competitors would have reacted to a decrease in demand for Peloton products, which could in turn affect the supply of Peloton products.” It would be different if the products had only one relevant attribute and that attribute was falsely advertised, since there the entire price paid would be based on falsity and no price premium analysis would be required. [The hypothetical is a joint pain cream that doesn’t cure joint pain; note that this is not a one-attribute product, since there are non-cream treatments for joint pain, so in fact there are at least two attributes, one of which is true in the hypo.]

The same flaws doomed plaintiffs’ omission theory, which relied on a consumer survey showing that 48.9% of respondents would be either extremely or moderately concerned if Peloton were forced by legal action to remove 50% of their classes. Peloton’s designated witness testified that there were only “a single-digit number of subscription cancellations that were directly attributable to ... the on-demand classes removed from [Peloton’s] library in March 2019.”