The folks suing Facebook for inflating video views now say it’s worse than they thought. Originally, the group said FB was jacking up video metrics by as much as 80% higher than they actually were by the way they were reporting them. Now, they say the real number is somewhere between 150% and 900%!
The social media platform admitted an error in the way it was reporting video accounted for over-reporting, but said the error did not affect customer billing. The group of marketers sued nonetheless, saying that the misrepresentation led them to believer FB’s video ads were more valuable than they actually were (based on the reach) and that drove prices up.
With that revelation, and other information, the marketing groups behind the lawsuit are now intimating outright fraud, according to reports in MediaPost. The complaint said that FB knew about the problems with the numbers in 2015, but didn’t acknowledge the problem publicly until a year later.
“Facebook engineers knew for over a year, and multiple advertisers had reported aberrant results caused by the miscalculation (such as 100% average watch times for their video ads). Yet Facebook did nothing to stop its dissemination of false metrics” – Excerpt from complaint (MediaPost)
This all started when one of the world’s largest advertising agencies, Publicis, announced that Facebook had told them the average time spent watching videos was overestimated between 60% and 80%, according to The Wall Street Journal. I’ll bet they’re not happy since they spent billions of dollars in ads.
What Facebook had admitted to was inflating the average duration of video views. A video view on Facebook doesn’t count until it plays at least three seconds. That’s better than what the Media Rating Council advocates – which is a view when at least 50% of the ad is visible on screen for at least two seconds.
What that means for duration though is that anything less than three seconds isn’t accounted for in Facebook’s duration metrics.
At the time, Facebook did apologize. “We sincerely apologize for the issues this has created for our clients. This error should not stand in the way of our ultimate goal, which is to do what’s in the best interest of our partners and their business growth. We can only be successful if we’re providing clients with the tools to drive their business forward, and we’ll continue to deliver on that promise,” said David Fischer – VP, Business and Marketing Partnerships.
Video is the future
“We see a world where video is first, with video at the heart of all of our apps and services,” Mark Zuckerberg, Facebook’s CEO in an investor conference.
So it’s probably important that Facebook gets its measurements right.
Facebook could have easily just update the metric stats and not said a word. Would people have noticed? Maybe, maybe not. FB is always tinkering with its algorithms and it has an impact on the metrics. Just ask publishers who have seen a continued decrease in referral traffic over the past years – in some cases dramatic.
If the social media company came clean on its own, cleaned up its problem, and made things right, then so be it. If they hid it from the people buying ads when they knew the data was bad, that’s a completely different story. And, if there was fraud involved, that’s an even bigger issue.
Facebook’s been catching it from all side lately. Another bug showed Facebook over-estimating unique reach by as much as 55%.
Facebook claimed it reached more people in the U.S. than actually live in the U.S. It’s pretty startling when you realize, according to a report from the Video Advertising Bureau, that Facebook claims to reach 39 million U.S. adults age 18-24… when there are only 31 million that live here – a discrepancy of roughly 8 million.
“It’s difficult to understand how a precision platform, such as Facebook, could continue to miscalculate these numbers time and time again. Rather, there appears to be a systematic misrepresentation of data across the board, at a scale unlike anything we’ve ever seen.” – Video Advertising Bureau via AdWeek
Facebook also reports 58 million potential reach of adults 25-34. That would be a pretty neat trick since there are only 44 million of them according to the US Census.
Digging into the data deeper though, you get a sense of how off these numbers look. Again, according to the Video Advertising Bureau, Facebook claims they can reach more people in every state than actual live there.
And it’s no different in the country’s largest cities.
“Whether this is truly another metrics glitch remains to be seen. However, with questions of trust regarding ad-tech platforms at an all-time high among many marketers, our analysis provides another instance where first-party data should at least be questioned, or even challenged, particularly when the numbers don’t align with universally accepted metrics such as U.S. Census Bureau population data and basic media math.” – Video Advertising Bureau