Two months ago we reported that in what may be just the first cockroach surrounding the massively overhyped advertising platform "potential" of Facebook, which for years has been accused of abusing ad clickfarms to inflate its ad metrics, big ad buyers and marketers were upset with Facebook after learning the tech giant vastly overestimated average viewing time for video ads on its platform for two years.
Back then Facebook admitted in a post on its “Advertiser Help Center” that its metric for the average time users spent watching videos was artificially inflated because it was only factoring in video views of more than three seconds. For the past two years Facebook only counted video views of more than three seconds when calculating its “Average Duration of Video Viewed” metric. Video views of under three seconds were not factored in, thereby inflating the average. Facebook’s new metric, “Average Watch Time”, will reflect video views of any duration. The WSJ calculated that ad buying agency Publicis Media was told by Facebook that the earlier counting method likely overestimated average time spent watching videos by between 60% and 80%.
Fast forward to today when it turns out that more cockroaches have indeed emerged after Facebook said it had "discovered" more mistakes in its reporting of ad-metrics to advertisers and vowed to be more transparent about errors in the future as it fixes the issue. This time, Facebook said it has found four other instances where it miscalculated reach on its site, including overstating how long people spent reading Instant Articles and how many people interacted with businesses' Facebook Pages.
According to Bloomberg, Facebook makes more than 220 metrics available to clients. Among the wrong numbers was one for "referrals" that was meant to show app makers how many of their posts directed traffic back to their website or app. Instead, Facebook's metric counted all the traffic that came out of the posts, even those that kept the viewer within Facebook. As a result, referrals were overstated by 6 percent on average, Facebook said.
In other cases, Facebook's math formula was simply wrong. For Instant Articles, which publishers use to help their stories load faster, Facebook said it had accidentally overstated time spent on stories by 7-8 % according to the social media. It had been showing clients a number that was an average over time, instead of just dividing time spent on stories by their number of views.
But most troubling disclosure was that For Facebook Pages, the company gave a number of how many people were reached, but double-counted repeat visitors. After the problem is fixed, pages may see their 28-day reach drop by 55 percent, the company said. More from the release:
one summary number showing 7-day or 28-day organic page reach was miscalculated as a simple sum of daily reach instead of de-duplicating repeat visitors over those periods (see red circle in screenshot below). However, the vast majority of reach data in the Page Insights dashboard (marked green in screenshots below) was unaffected, including all the graphs, daily and historical reach, per-post reach, exported and API reach data, and all data on the Reach tab. The de-duplicated 7-day summary in the overview dashboard will be 33% lower on average and 28-day will be 55% lower; data in other fields is unaffected. This bug has been live since May; we will be fixing this in the next few weeks. It does not affect paid reach.
Needless to say, companies and marketers rely on Facebook to be honest and report to them how well the content they post is performing, so that they can make strategic decisions about what to do next and how much to invest through advertising - clearly, the "better" the performance, and the more overstated Facebook's influence, the greater the ad budget allocated to FaceBook.
To avoid future errors, the company said it would establish a measurement council made up of top advertisers and partners. But much more importantly, the company said it would also allow more third party measurement companies such as Nielsen to track and supplement its metrics. Additionally, Facebook is revising the descriptions for its data to explain exactly what they measure, for example reporting "3-second video views" instead of just "video views."
To be sure, all welcome remedial moves by the social network - whose big crusade these days is eliminating "fake news" - and yet with two visible incidents both of which "oddly" swayed the erroneous data in Facebook's favor in two months, which forced Facebook to go so far as accepting third party supervision over its core business, which is not page views but trust, one wonders if some of Facebook's clients haven't had enough, especially in light of the social network's prominent and controversial role in the recent presidential elections.
Finally, Facebook said it would disclose further updates in the coming weeks on a new blog that's specifically dedicated to metrics news. Facebook said none of the issues would affect client billing. "We strive to be as accurate as possible," Carolyn Everson, Facebook's vice president of global marketing solutions, told Bloomberg in an interview. "I can't sit here and tell you it's going to be perfect every day, because frankly, consumer behavior changes too quickly." So it was all the fickle consumer's fault?