Antitrust Law Daily Wrap Up, CONSUMER PROTECTION—W.D. Wash.: FTC arguments to exclude Amazon expert witness testimony went to weight, not admissibility, (Aug 7, 2025)

By Justin Marcus Smith, J.D.

The selection of an inappropriate survey universe generally affects the weight of resulting survey data.

Purported technical defects with an Amazon expert’s survey of Amazon users did not warrant exclusion of expert testimony about the survey results, held the federal district court in Seattle, Washington. The court suggested the FTC could attack Amazon’s expert evidence by cross examination, contrary evidence, and attention to the burden of proof, but not by exclusion. The court also found a second survey relevant to FTC allegations about “dark pattern” consumer engagement with online disclosures in connection with the cancellation of Amazon Prime trial memberships. The court denied the FTC motion to exclude the Amazon expert’s testimony about the two surveys (FTC v. Amazon.com, Inc. , No. 2:23-cv-00932-JHC (W.D. Wash. Aug. 6, 2025)).

Background. The FTC sued Amazon and three Amazon executives for alleged violation of Section 5(a) of the FTC Act and Section 4 of the Restore Online Shoppers’ Confidence Act (ROSCA). The FTC contended Amazon did not clearly and conspicuously disclose its cancellation process. The FTC also alleged reasonable consumers might not know they were consenting to sign up for Prime subscription auto-renewal.

In May 2024, the court denied Amazon.com motions to dismiss because the court found the Amended Complaint pleaded sufficient facts. In November 2024, the court also denied an FTC motion to strike three Amazon affirmative defenses: (1) unclean hands; (2) equitable estoppel; and (3) laches.

In February 2025, the court denied an Amazon motion to compel a deposition about public statements the FTC made about the Restore Online Shoppers’ Confidence Act (ROSCA) during Negative Option Rulemaking. The court found the Amazon the FTC suit had to do with ROSCA, not the Negative Option Rule; therefore, Amazon was seeking facially irrelevant testimony.

Proposed expert. Amazon’s proposed expert is a doctoral marketing professor at the University of Virginia’s Darden Graduate School of Business Administration. His purported areas of expertise within marketing include branding, consumer behavior, surveys, statistical modeling of consumer choice, and the public policy implications of marketing.

Two surveys. Amazon asked the Darden expert to conduct two surveys: (1) a survey to assess which Amazon Prime members can locate and complete the desktop version of the Amazon.com Prime membership cancellation flow, as the FTC described it in the amended complaint (cancellation survey); (2) a survey to assess U.S. consumer experiences with free memberships or subscriptions that automatically turn into paid memberships if not canceled before the end of the free trial period (free trial survey). Amazon said the expert relied on his training, experience, and expertise in marketing and consumer behavior. It said he also considered academic literature and case materials.

Cancellation survey. The expert reported that the cancellation survey showed 99.8% of respondents located the cancellation flow and 96.4% paused or ended their Prime membership. The cancellation process purportedly took an average of 74 seconds. These results were obviously inconsistent with the FTC allegation that the process inhibited cancellation.

Free trial survey. The free trial survey purportedly found that 92.% of respondents were currently paying for at least one membership or subscription; 58.0% reported signing up for one or more free trial memberships in the last 12 months; and 85% said their most recent free trial automatically turned into a paid membership or would have done so if they did not cancel it before the end of the trial period.

FTC argument. Essentially, the FTC sought to overcome the high bar to exclude the survey evidence for flawed methodology and irrelevance. The FTC articulated two reasons for excluding testimony about the purported survey results: (1) the expert recruited and sampled only the most tech-savvy survey takers; and (2) the expert failed to ensure his results were representative of U.S. Amazon Prime members who sought to cancel their memberships. In addition, the FTC said the second, free trial survey had nothing to do with Prime memberships.

Analysis. The court agreed with Amazon that testimony about the cancellation survey was admissible. The FTC claimed there was selection bias toward tech-savvy and attentive survey takers who were not representative of U.S. Prime members. In addition, the FTC argued the expert did not verify key demographic representation among respondents.

Amazon countered that FTC criticisms went to weight, not admissibility. Amazon also said the FTC was attempting to reframe a sampling argument into a general attack on whether the expert followed accepted survey principles without citing any caselaw to support that as a distinct attack.

The court found the expert followed generally accepted survey guidelines in applying his own purported experience and expertise in conducting numerous surveys. Following generally accepted survey techniques would made the survey sufficiently reliable. The focus on the cancellation flow and whether respondents could cancel their Prime memberships was relevant to the FTC case.

The court agreed with Amazon that sampling criticisms were an attack on survey design and methodology. They therefore went to weight, not admissibility. The expert’s use of attention checks also went to weight because, again, they were about purported technical or methodological unreliability. The third FTC point about the representation of the general population of Prime members also went to weight, as did the fourth FTC point about socio-demographic information or representation.

The court found the second, free trial survey engaged with the FTC argument that context matters when evaluating the conspicuousness of online disclosures. The free trial survey was related to the FTC argument that the alleged Amazon use of “dark pattern” online user interface strategies made it unlikely that many ordinary consumers would notice enrollment in free Prime trial membership. The court found the survey purported to show general awareness of auto-renewal. The court held the free trial survey was relevant and admissible because it might help the trier of fact.

The Case is No. 2:23-cv-00932-JHC.

Judge: Chun, J.

Attorneys: Evan Mendelson for the FTC. Anders Linderot (Covington & Burling LLP) for Amazon.com Inc.

Companies: Amazon.com Inc.

Cases: ConsumerProtection WashingtonNews FederalTradeCommissionNews