Just Like TV, Akamai Research Says Those Ads in the Middle of Video Do Get Seen


Apparently the idea of capturing a more or less captive audience—something TV does with every commercial spliced into the middle of programming—is a pretty effective way to keep people watching.  

Akami Technologies tomorrow will formally unveil its new study, Understanding the Effectiveness of Video Ads: A Measurement Study, coauthored by S.S. Krishnan and R.K. Sitaraman, at a confab in Barcelona.
It shows that an advertising message online gets seen to completion most in the middle of a video than after the video. Pre-roll, the most familiar spot, seems to be an adequate compromise .

Mid-roll adsare 18.1% more likely to be completed than the same ad as a pre-roll, and 14.3% more likely as a pre-roll than as a post-roll. I can’t really recall ever seeing a post-roll ad, which means I prove the validity of that finding, for sure.

But while I'd be likely to stick around for a mid-roll ad to finish, I'm not so sure I wouldn't make a mental note to avoid that product. I think they leave a bad impression --but it is an impression-- even in longer videos which is where they usually show up.  

The study analyzed in aggregate 367 million videos containing 257 million ads from all over the world, from 3,000 publishers and viewed by 65 million unique users. Nothing of the research is likely to make you do a spit-take, so feel free to snack while reading.

It is probably not surprising that viewers seem more tolerant of video ads than they are of slow-loading videos, which is kind of like a comparing a rotten apple to a frozen orange.

Viewers who have wait 10 seconds or more for a video to begin are three times as likely to abandon the whole idea that online viewers who have to spend the same time watching a pre-roll advertisement.

And this just in: Users who don’t want to watch pre-roll make their feelings known pretty darn quickly. One third leave at or before even 25% of the ad has been unspooled and two-thirds are gone before the halfway point.  

So, that adds stats to my sympathies for pre-roll advertisers. When you lose a third within split seconds, that means you have to have one helluva instant hook. I always like the ones that start: “Attention consumers!”     

On the plus side, the research discloses that loyalty to publishers does mean something. Repeat visitors to a site have much higher completion rates than one time drive-by online guests, suggesting there is some level of trust viewers associates with sites they know about. Either that or for whatever reason, they just need the information the site presents, and are compelled to watch the ad.     

1 comment about "Just Like TV, Akamai Research Says Those Ads in the Middle of Video Do Get Seen".
Check to receive email when comments are posted.
  1. John Grono from GAP Research, October 23, 2013 at 7:46 p.m.

    It appears as though this research was conducted server-side and not consumer-side. So, sure the mid-roll may have 'completed' (i.e. no viewer clicks and interruptions), but that makes it a "viewable" ad rather than a "viewed" ad. For example, the user may minimise the browser that the player is in and use another application, or move the focus to a different tab in that active browser (in essence, they do something else that they deem more productive), and then when the ad audio ends return the focus back to the browser tab with the video player. Meanwhile back on the server the ad video is still streaming away merrily thinking someone is watching it.

Next story loading loading..