Facebook's bad: Company clarifies stand on video metric fiasco

Last week, Facebook confessed to advertisers what they and the ad world knew all along about videos on the platform. That the social network giants’ video metrics are not as clear cut as many would like.
Here’s the disclosure, clarified again by Facebook’s VP, business and marketing partnerships, David Fischer: “About a month ago, we found an error in the way we calculate one of the video metrics on our dashboard – average duration of video viewed. The metric should have reflected the total time spent watching a video divided by the total number of people who played the video. But it didn’t – it reflected the
total time spent watching a video divided by only the number of “views” of a video (that is, when the video was watched for three or more
seconds). And so the miscalculation overstated this metric. While this is only one of the many metrics marketers look at, we take any mistake
seriously.”

Facebook’s standard reply to the press on the issue was: “This error has been fixed, it did not impact billing, and we have notified our partners both through our product dashboards and via sales and publisher outreach. We also renamed the metric to make it clearer what we
measure. This metric is one of many our partners use to assess their video campaigns.” What was called “Average Duration of Video
Watch Time”.

But that wasn’t enough to assuage the concerns of agencies and advertisers. Some of whose thoughts we’ve highlighted here. The bottom line, however, is that nothing really changes. As Saurabh Kanwar, the founder of Flarepath, says, without mincing words, “Nobody is really giving a f***.” Except, he points out, the case for YouTube just got
stronger.

Facebook’s ‘My Bad!’
Well, what do you expect from someone who auto-plays videos and forces views? Unfortunately, nothing will come off it unless advertisers do a class action suit against them which they won’t given the incestuous nature of the relationship. Facebook members don’t care and are too addicted to bother about such stuff. At least YouTube calculates a view after 10 seconds, which is fair. My content guys say that they had figured this out anyway and knew that the numbers are skewed.

Sachin Bhatia
CEO and co-founder of dating app, TrulyMadly

The key issue with Facebook videos has been the rampant over-reporting of video views in an effort to catch up with Google’s YouTube. While YouTube counts a video as viewed only when it’s seen in full, per its TruView numbers, Facebook would count its videos as seen by
playing it live on loading the page and then billing clients for them. Even though the vast majority of views were of the “less than 3 seconds” variety, they were charged for in full. The new incident is a compounding of that misinformation.

“Average Watch Time” is a more accurate metric. And more challenging for creative teams to produce videos that are viewed for longer.
The strange thing is that clients haven’t cared. Because agencies haven’t cared. And agencies haven’t cared because they got to espouse Facebook videos more, and by doing so to bill more, and hence to collect more commission from clients.

YouTube’s TruView set the gold standard here in transparency. But Facebook Videos continues to play fast and loose with numbers and definitions. Given that agencies seem to work more for Facebook and less for their clients, this lack of integrity has regrettably not made a difference.

Mahesh Murthy
Venture capitalist and founder of digital marketing consulting firm Pinstorm

The key metric for any media platform is watch-time and engagement. Views is a reach metric, and while it remains a necessary ingredient,
its definition is open to interpretation. Watch time on the other hand has a universal language, and is already the standard for both YouTube and NetFlix.