In a post on its advertising help center, a Facebook employee announced the discrepancy and explained the difference between how it defined the statistic, and how it was actually measured.
We had previously *defined* the Average Duration of Video Viewed as "total time spent watching a video divided by the total number of people who have played the video." But we erroneously had *calculated* the Average Duration of Video Viewed as "the total time spent watching a video divided by *only* the number of people who have viewed a video for three or more seconds."
In response, Facebook says it's introducing two new metrics:
Video Average Watch Time: the total watch time for your video, divided by the total number of video plays. This includes plays that start automatically and on click. This will replace the Average Duration of Video Viewed metric.
Video Percentage Watched: reflects the percentage of your video somebody watches per session, averaged across all sessions of your video where the video auto-played or was clicked to play. This will replace the Average % Video Viewed metric.
As a user, this probably doesn't affect you much. But even though Facebook says the discrepancy didn't affect billing, advertisers who relied on the numbers and outlets (like Engadget) who posted video to the platform
may have more questions. Bloomberg
points out that Facebook is set to meet top advertisers next week during the Advertiser Week conference -- we probably haven't heard the last of this.