Rhythm NewMedia recently released their Mobile Video Advertising Report for Q4 2010 which had some interesting findings in it. However, they are the largest mobile video ad network and all of the findings in the report are based on metrics from their network so a grain of salt is required here I think.
Here is some information directly from their report:
Data points are based on ads served across iPhone, iPod Touch, iPad, Android and other devices in the US market via Rhythm NewMedia's network. Rhythm's publisher network includes only trusted apps and mobile websites including top broadcast TV networks, 10 of the top 25 cable networks, the largest news agency, and leading entertainment, news, games, sports, lifestyle, music sites and more. Rhythm's network does not include any user generated content. The data points cover ad campaigns from over 200 Fortune 500 brands this year, with over 70 brands in Q4 alone.
Now that I hope you're considering the source of the information in the proper light, let's take a look at what they found. They compared their data against that of Alex Mindlin's New York Times article, Short Attention Spans for Web Videos. An odd article title considering that comScore just reported the average length of videos watched online is 5 minutes.
In the article Mindlin reported that 19.4% of video viewers abandon online video after 10 second and 44.1% abandon the video after one minute according to research from Visible Measure.
In comparison, Rhythm NewMedia reports that mobile video is 13% stickier for the first ten seconds and a full 25% stickier after one minute.
Those are some pretty high numbers to say the least. I find them quite hard to believe considering that most mobile video is being watched on a relatively small screen, with somewhat slow network speeds.
Then again, they are including the iPad in there which I don't know is exactly mobile video. Sales of the iPad have reportedly topped 9 million units (though Gartner estimated 19 million and others 21M) and there are a reported 153 million mobile phone subscribers, so it could skew the data somewhat. Of all iPad sales only about 58% of them are 3G so even then, only that percentage of iPads should be considered 'mobile' devices and the rest should be considered portable computers, since that's what they are. In fact, some might even argue that since you can't make calls on the iPad it's not a mobile device at all.
They also looked at video completion rates in straight mobile web videos versus apps and found that viewers only watch 2 videos each on the web versus 11 videos per viewer in apps.
The numbers are certainly of some interest. It seems that, according to their research, mobile users aren't all that interested in watching videos via their web browsers but prefer to watch them via specific applications. There are numerous apps for the major broadcasters, YouTube and other video sites so it's no surprise that people would prefer video-specific apps versus the more generic web browser which, especially on the iPhone, doesn't handle video as well.
What I would like to see is a third-party analysis of data across multiple mobile video ad networks so that we can get a clearer picture of the mobile video landscape. Included should also be a breakdown by device or at least by browser or application which would go a long way toward giving us all some vital information into the mobile video landscape. I would also like to see some specific breakdown on both interactivity and engagement. Publishing a report based on your own internal numbers and then comparing it to the numbers of others seems more like a marketing tactic for your business and less like actual research to me.