DZ Reserve v. Meta Platforms, Inc.

Background

DZ Reserve and Cain Maxwell sued Meta Platforms, Inc., formerly known as Facebook, alleging fraud involving Meta’s advertising services. The plaintiffs claimed that Meta’s “Potential Reach” metric overstated the number of people an advertisement could reach and caused advertisers to pay artificially high premiums for ad placements.

The court said three claims remained relevant to class certification: fraudulent misrepresentation, fraudulent concealment, and a California Unfair Competition Law claim seeking injunctive relief. The court had already dismissed the plaintiffs’ claims for breach of the implied covenant of good faith and fair dealing, quasi-contract, and restitution under the California Unfair Competition Law. The fraud claims were limited to conduct occurring on or after August 15, 2015.

Proposed Classes and Certification Analysis

The proposed class covered all United States residents, including individuals and incorporated entities, who from August 15, 2014, through the present paid for at least one advertisement on Facebook’s platforms through Facebook’s Ads Manager or Power Editor. The proposed definition excluded several categories of advertisements, including advertisements bought under agreements other than Meta’s specified terms, advertisements using only non-lookalike Custom Audiences, Reach and Frequency purchases, specified advertising objectives, and advertisements for which Meta provided a Potential Reach below 1,000.

The court concluded that the proposed class satisfied the requirements for certification under Federal Rule of Civil Procedure 23. It found numerosity because more than two million U.S. advertisers purchased Facebook ads during each year of the class period. It also found that the named plaintiffs’ claims were typical of the class and that they and their counsel could adequately represent the class. The court rejected Meta’s argument that differences between large corporations, small businesses, and individual advertisers defeated typicality or adequacy.

For commonality and predominance, the court identified the central common question as whether Meta’s Potential Reach metric misled advertisers. The court found that common evidence could address whether Meta represented Potential Reach as the number of people an advertisement could reach even though the metric estimated accounts, whether Meta knew of the discrepancy, whether it intended advertisers to rely on the metric, whether the representation was material, and whether advertisers suffered injury.

The court also found that damages could be calculated on a classwide basis. The plaintiffs’ experts used analyses concerning inflation in Potential Reach, the effect of that inflation on advertising budgets, and resulting price premiums. The court determined that the damages methodology did not need mathematical precision but had to provide a reasonably accurate classwide measure tied to the plaintiffs’ theory of liability. It further found that proceeding as a class was superior to individual lawsuits because the amount at issue for each advertiser was relatively small compared with the resources required to litigate individually.

For the California Unfair Competition Law claim, the court certified a class seeking injunctive relief under Rule 23(b)(2). The proposed injunction would require Meta either to correct the Potential Reach metric by removing known sources of inflation or to remove the metric. The court found that the plaintiffs had standing to seek an injunction because they testified that they would consider buying Meta advertisements again if Meta corrected or removed the misleading metric.

Expert Motions and Disposition

Under Federal Rule of Evidence 702 and the standard established in Daubert v. Merrell Dow Pharmaceuticals, Inc., the court evaluated whether the challenged expert testimony was reliable and useful at the class-certification stage.

The court denied Meta’s motion to exclude Dr. Allenby’s report and testimony. Dr. Allenby used a conjoint survey and statistical analysis to assess the effect of inflated Potential Reach on advertisers’ budgets. The court found that conjoint analysis and linear regression were accepted methods and concluded that Meta’s criticisms generally concerned the weight of the opinion rather than its admissibility.

The court granted Meta’s motion to exclude Mr. McFarlane’s report and testimony. It found that his report did not provide specialized or scientific expertise beyond the ordinary knowledge and experience of a jury. The court also excluded any part of Dr. Roughgarden’s opinions drawn from Mr. McFarlane’s work unless an independent basis could later be shown; the court left that issue open for a possible motion before trial.

Order

The court certified the proposed class under Rule 23(b)(3) for the common-law fraud claims and under Rule 23(b)(2) for the California Unfair Competition Law injunction claim. DZ Reserve, Inc. and Cain Maxwell were appointed class representatives, and their counsel were appointed class counsel. The plaintiffs were directed to submit a proposed notice-distribution plan by April 29, 2022. The court also set a status conference, directed the parties to propose dates for the final pretrial conference and trial, and referred the parties to Magistrate Judge Hixson for a settlement conference.