Imagine this: you’re just trying to watch a nice relaxing video of a 14 year old white boy hitting himself in the crotch with his own skateboard, when all of a sudden a pre-roll ad pops up in front of the content you’re viewing. Sucks, right? So researchers at the University of Massachusetts Amherst decided to calculate just how much it sucks.
According to the oodles of research that computer scientists Ramesh Sitaraman and S. Shunmuga Krishnan of Akamai compiled, when an advertisement is inserted into middle of the video, viewers are more likely to pay attention to it the whole way through — these types of ads have a 97 percent completion rate, whereas ads inserted at the beginning have a 74% rate and ads at the end, 45% . This makes an odd sort of sense, as the majority of us grew up watching television and are pretty accustomed to having our content interrupted by advertisements part of the way through.
However, other factors contribute to whether or not someone will watch an ad all the way through. When the team used an intentionally slow-loading video, people jumped ship almost immediately; after ten seconds, more than 45% of people abandoned the slow ad. They’re also much less willing to sit through an ad if the content they’re hoping to get through is shorter, like a news clip or a quick cat video. And viewers who repeatedly come to one particular site for video content are actually more willing to watch the ads — perhaps because they’re the exact type of person who the advertisement is geared towards, or maybe just because they feel like they have to pay it forward for a site they really like.
This is reportedly the largest scale scientific study of video ad efficacy that has ever been attempted, using anonymous viewer data that’s been linked to more than 257 million ads, which were then inserted into 367 million videos from over 3,000 publishers and viewed by 65 million unique viewers globally. And all this data was collected in only a 15 day period in April of this year. Wow, the Internet is big. I mean, really big. Vastly, hugely, mindbogglingly big.
Sitaraman plans to present the bulk of their findings ACM Internet Measurement Conference in Barcelona on October 24. He says that he hopes learning more about to to make “online videos economically sustainable and profitable” will contribute to the economic future of the Inernet.
(via Phys.org)
- Twitter got in trouble for creating fake tweets to publicize their new ad platform
- Apple TV might be trying to get rid of ads entirely
- Even the guy who invented banner ads hates banner ads
