A group of small advertisers filed a lawsuit in California federal court in 2016, alleging the tech giant engaged in unfair business conduct by disseminating inaccurate metrics that significantly overestimated the amount of time users were spending watching video ads.
The plaintiffs later added a fraud claim, and in Tuesday’s court filing they alleged Facebook knew of irregularities in its video metrics by January 2015 and understood the nature of the miscalculation within a few months, but failed to disclose the information for over a year.
The filing followed the plaintiffs’ review of some 80,000 pages of internal Facebook records that they obtained as part of court proceedings.
The complaint, which cites the internal Facebook documents, also alleges that the scale of the miscalculation was far worse than understood.
“Facebook’s internal efforts behind the scenes reflect a company mentality of reckless indifference toward the accuracy of its metrics,” the plaintiffs said in Tuesday’s filing.
In a statement, a Facebook spokeswoman said, “Suggestions that we in any way tried to hide this issue from our partners are false. We told our customers about the error when we discovered it—and updated our help center to explain the issue.”
Facebook said the lawsuit is without merit and has moved to dismiss the fraud claim.
The plaintiffs in the case include Crowd Siren, a small Las Vegas marketing agency, and Jonathan Murdough, a Pennsylvania resident who purchased Facebook video ads.
The lawsuit, which seeks class-action status and punitive damages, stemmed from a September 2016 Wall Street Journal report that said Facebook had vastly overestimated average viewing time for video ads. Facebook disclosed the issue in a post on its advertiser help center that August.
Facebook replaced the metric with “average watch time,” which reflects video views of any duration.
After disclosing the issue in 2016, Facebook said in a statement that it had “recently discovered” the error.
Facebook told some advertisers that it likely overestimated average time spent watching videos by 60% to 80%. The plaintiffs alleged in Tuesday’s complaint that the error was much larger and that the average viewership metrics had been inflated by some 150% to 900%.
Facebook also said at the time that the error didn’t affect billings. However, in their complaint, the plaintiffs claim Facebook’s misrepresentations “induced” advertisers to purchase video ads and to pay more for Facebook’s video ads because they believed users were watching videos for longer than they actually were on average.
The claims that Facebook failed to act when it discovered the video metric error were in an August filing, but were heavily redacted at the time. In the latest version of the complaint filed Tuesday, those claims were unredacted.
The plaintiffs allege the Facebook documents show that by July 2015 the company had received inquiries from several advertisers about video metrics that appeared suspect, and had essentially determined the cause of the issue.
In June 2016, nearly a year later, a Facebook engineering manager, following up on advertisers’ complaints, discussed the issue internally, writing, “somehow there was no progress on the task for the year.” The plaintiffs also allege the company developed a “no PR” strategy to avoid drawing attention to the matter.
Facebook decided to “obfuscate the fact that we screwed up the math,” the complaint said, quoting the Facebook documents.
After the video-metric error, Facebook disclosed other errors in its measurement practices on several occasions and came under renewed pressure from the ad industry to make changes.
The Association of National Advertisers, a trade group that represents top marketers including Procter & Gamble Co., General Electric Co. and Verizon Communications Inc., called on the company and other major digital ad-sellers to allow independent verification of their metrics.
Facebook has worked to address the matter by allowing more third-party measurement companies to validate its data, and by undergoing audits by the Media Rating Council, the media industry’s measurement watchdog.
Jason Kint, chief executive of Digital Content Next, a trade organization that represents online publishers, said Facebook’s measurement practices have been detrimental to the digital advertising marketplace. “Facebook needs to lead with radical marketer and consumer transparency to get past this. We haven’t seen it yet,” he said.
The error, and Facebook’s handling of it, became a critical moment in the relationship between the social-media giant and marketers who pay its bills. Many brands already were skeptical of the practice by Facebook and other tech giants to closely guard their internal ad data—one top executive likened it to “grading their own homework.” The incident fueled renewed calls for Facebook to allow independent measurement and auditing.
Damping marketers’ concerns is important for Facebook as it seeks a bigger piece of U.S. spending on online video ads, which is projected to grow 30% this year to $27.8 billion. Facebook is expected to account for almost 25% of U.S. video ad spending, eMarketer estimates.
For two years, Facebook had counted only video views that lasted more than three seconds when calculating its “average duration of video viewed” metric. Video views of under three seconds weren’t factored in, thereby inflating the average length of a view.