I've been doing a lot of asking and listening over the past few months, meeting with both advertisers and publishers to learn their priorities and needs when it comes to video. In those conversations, several common threads have emerged. Here they are below.
Pricing is a struggle. Publishers are understandably attempting to maintain premium pricing for their in-stream video advertising units. Many, particularly in the back half of the year, have delivered substantial sell-through at healthy CPMs. Some have not. Marketers have largely focused on the disparity between TV CPMs and their online equivalent, with the latter being much higher than the former. Measurability and interactivity don't seem enough to overcome this disparity. Interestingly, rich-media formats seem to be well positioned to benefit from this disparity, with in-banner formats (at least anecdotally) enjoying an uptick due to their cost efficiencies and scale.
Quality scale is (still) lacking. Despite increased online content and viewership, there still appears to be an imbalance of supply and demand for top quality in-stream advertising opportunities. Hulu and YouTube aside, many publishers bemoan their lack of available inventory to accommodate the larger RFP sizes that tend to be attached to video. Savvy publishers are exploring syndication and audience extension partnerships with ad networks to bring more opportunity to the table. Short-term, this is a great strategy. However, with video spending forecast to grow 40% per year for the next few years, the supply issue (combined with the pricing struggle noted above) may hamper that growth trajectory.
Metrics and measurement are a focus. Cost-per-engagement (CPE) was an interesting topic a year or so ago, but failed to become a trend. With the proliferation of more in-banner opportunities, CPE has made a bit of a comeback. Additionally, as ad serving capabilities have grown, a push toward evaluating -- and paying for -- verified and viewed advertisements is growing. Witness YouTube's recent announcement of testing "skippable" pre-roll, in which advertisers will only pay for viewed impressions. Experimental models such as this are a very positive development for a growing industry, and may partially address some of the above two issues.
Those are three key issues of the day as we close out the year. Please add your own in the comments. Next month, we'll tackle 2010 online video predictions.