Over the years, online video content has become an extremely valuable tool for advertisers to create excitement and buzz around a brand or product launch. In addition to a company’s owned media channels, platforms like YouTube and Vimeo have dominated this space for years. Facebook however, has only recently taken steps to become a significant player in this space with its video hosting services.
Facebook has had the ability to host videos for quite some time, however in an effort to establish itself further in video space it recently introduced the “auto-play” feature for each user, causing a significant increase in engagement rates for videos.
While most advertisers have been pleasantly surprised by this change, the actual results can be misleading. Digging a little deeper into the overall video performance details, we notice that videos performed better before auto-play was introduced.
When the auto-play feature was activated, Facebook set the minimum viewing time needed to be counted as an engagement to three seconds. This change allowed Facebook to quantify any video playing for three seconds as an engagement, whether the user activated sound or if the video played on mute while simply scrolling their newsfeed. Therefore, this feature presents a very misleading engagement rate.
The graph below compares how much of the video users are actually watching before and after the auto-play introduction. This data come from a sample of our client’s video ads campaigns.
As we can see, there is a significant difference across the board between the two. This difference is expected as users need to actively click to play the video before auto-play was introduced. Previously, the engagement numbers suggested that there was a clear intent and interest from the user to engage with the video. Now that auto-play is active, users are less likely to actively engage and watch videos that automatically begin playing.
Since introducing the auto-play feature, Facebook developed the possibility of creating audience lists from the users that engaged with the video. These lists specifically allow advertisers to categorize users based on the different video view percentages; completed (95% and above) and non-completed (50% and below).
Once these lists are generated, advertisers can leverage it and create look-a-like audiences into the next video campaign, which will likely increase the upcoming campaign’s overall video view percentage.
As advertisers, it is important to understand the subtle differences between each platform, what they offer, and how they connect with the targeted audience. As Facebook continues to grow and mature as a video content platform, a significant focus will need to be placed on the platform’s true engagement levels in order to determine its value compared to its competitors.
This article was originally written by Matthew Gangnier.