Facebook, the world’s social media powerhouse, has admitted and apologised for recent video metric miscalculation that led marketers to overestimate how much time people actually spend watching video ads on the platform.
It was recently revealed that for the past two years, Facebook has been overstating video view times by 60 to 80 percent. The miscalculation occurred because the platform had only been taking into account views that last more than three seconds when calculating its ‘Average Duration of Video Viewed’ metric instead of total views.
Until Facebook can demonstrate its figures match up reliably to third-party analysis, marketers should proceed with caution.
The error has now been fixed, according to a deeply apologetic Facebook. However, this news is likely to leave marketers second-guessing their video advert spend at a time when Facebook is touting video more than ever before.
Facebook has now stressed that this error is another reason why the company is increasingly relying on third-party verification. (Just a day before the miscalculations went public, Facebook announced a number of partnerships with measurement companies.)
It’s all very well Facebook stating that it welcomes third-party verification. However, for small businesses and marketers that don’t have the budget for extensive third party verification systems, the news is concerning. Facebook has built an ads dashboard that provides marketers with extensive reports and analytics. Whilst large advertisers will use third party tools to cross reference results as a matter of course, I’d wager that the vast majority of SMEs rely on nothing but the Facebook ads dashboard. These companies have based their marketing budgets on figures provided by Facebook, perhaps choosing social as it seems that their money goes further than in other placements. If Facebook can’t provide robust and impartial analytics, I imagine those marketing budgets will be reallocated elsewhere.
If Facebook can’t provide robust and impartial analytics, I imagine marketing budgets will be reallocated elsewhere.
Whilst there have always been suspicions that Facebook massages its figures, the received wisdom has generally been that this is done via the way they’re presented, rather than outright miscalculation.
For example, it’s no secret that Facebook counts three second video views, likely to be nothing more than an autoplay as someone scrolls through their feed, as video views in ad campaigns, and charges accordingly.
However, marketers always felt they had the tools to distinguish these ‘accidental’ views from more valuable engagements: Facebook provided metrics on 30 second views, views to 95%, and so on. These revelations invalidate these figures. Whilst the video views issue has now been fixed, marketers will be wondering which other elements of their Facebook reports are affected by errors in Facebook’s favour.
Until Facebook can demonstrate its figures match up reliably to third-party analysis, marketers should proceed with caution. Brands and agencies that do not have the resource to verify their social marketing analytics using external tools should keep their ear to the ground for rumblings of other errors – or risk being caught out.
Do you use third party tools to verify Facebook analytics for your company? Do the results consistently match Facebook’s? I’d love to know. Comment below or tweet me @AgostinelliAldo.