By now, most of you already know how one company caused such a buzz in both the media research and online advertising spaces at the Advertising Research Foundation's third annual Audience Measurement Symposium, held on June 24 and 25. But for those of you who have been on vacation the past two weeks, of course I'm talking about my 10-minute slot on Tuesday morning's video panel, which I titled "Main Stream: Understanding and Measuring Online Video Consumption." Of course a lot of companies save their audience measurement news for this conference -- in fact, a guy I know told me that one of the search engines actually launched a research tool there. Maybe he's right; I confess that I was so absorbed in my own ten minutes that a lot of the other news just went in one ear and out the other. But if it happened, I'm sure you can find it somewhere in the MediaPost archives.
So where was I? Oh, yeah...
Each month, between 70% and 75% of online Americans watch online video. Not surprisingly, online video is a broadband phenomenon; over 99.5% of all online video in the U.S. is streamed over a broadband connection. It wasn't so very long ago that the prospect of streaming as a viable national medium was tied to the penetration of broadband; when I consulted for my friend Bill Rose at Arbitron circa 2002, we talked about how home broadband penetration of 30% would be the threshold required for streaming to be considered a truly national medium. According to comScore, about 85% of Internet users had access to a broadband connection at home as of April 2008 -- up about 40 points from four years prior. At work, broadband penetration is even higher.
We're used to the face of the Internet changing rapidly as new applications take root and spread; five years ago no one knew what "social networking" was, and if they did, it was because they were on Friendster. (And I'm old enough o remember when Pointcast was revolutionizing the Internet with "push.") So the tsunami that is online video seems almost to have crept up on us. From February 2007 to April 2008 -- 15 months -- monthly videos per U.S. online viewer have increased two thirds, while minutes per viewer have increased almost 80%. (Of course, the fact that total minutes have grown faster than number of videos watched means the average length of stream is increasing.) Obviously YouTube opened the floodgates -- in April, about 37% of videos watched in the U.S. came from YouTube, whose share is about double what it was a year ago. (Note that because of the generally short duration of a YouTube stream, its share of minutes is somewhat less; 32% as opposed to the 37%.)
I tend to think there are three general business models for online video. One is CGM --the YouTube model, Mentos in coke bottles or my cute kid in the bathtub. One is short-form professional video -- clips, highlights, the Letterman top-10 list. The last is professional long-form -- full TV programs, ballgames, movies. We've seen a lot of interest from both publishers and agencies around that last model. Some of my colleagues believe that the nature of the Internet is fundamentally different than TV, and that online users bring short attention spans, and are accustomed to surfing, to moving from one piece of brief content to another.
This school of thought suggests that it is short-form video that will continue to rule online video. I'm not convinced; the Internet is not just a medium (per se); it is also a distribution platform. In other words, it can be compared to TV, but it can also be compared to broadcast or cable or satellite (the technology that distributes TV.) I tend to believe that we'll continue to see Internet streaming eat into the distribution share of other video distribution technologies (broadcast, cable) as a means for consumers accessing long-form professional content. Arguing for this is the fact that the heaviest online video users are 18- to 24-year-olds and 25- to 34-year-olds; young people will continue to come of age in a world where pipelines and screens are interchangeable. In fact, I've heard anecdotally that many under-25s don't even own a TV, opting instead to stream programs that interest them, or waiting until after the season and renting the DVDs.
Of course it is also possible that in terms of sheer tonnage of usage, the CGM model will hold sway (email me for the link to the adorable video of my daughter in the bathtub.) But in terms of monetization -- of traditional advertising revenue, of putting commercials into video and selling that time -- long-form professional content still looks like the moneymaker to me. This may be sacrilege, but I'm still not convinced that the key to monetization online lies in the long tail.
One last point, and then I'm going to go finish this vacation I'm on. In the analog age, the broadcast age, TV audiences accumulated all at once, and even now, we assume they accumulate pretty close to all-at-once (how different is live from near-live from live-plus-three, and heck, three days isn't that long a window.) As the video distribution model changes from linear (this show on at this time) to on-demand (PPV, VOD, streaming), programs will accumulate their audience with a velocity approaching print audience accumulation. Air date will become more akin to newsstand date -- the time the content is released, and the beginning of audience accumulation.
Interestingly, at that ARF conference, another speaker talked about print content resembling broadcast as it migrates online. I guess, in this first column after July 4, I should close by calling the Internet the great melting pot, and leave it at that.