Long-form video (anything longer than 30 seconds) can be distributed using a variety of mechanisms. As most industry folks are aware, generating incremental audience for these videos has largely revolved around structuring revenue share arrangements with third-party publishers. Under this arrangement, third-party sites sell the advertising avails and split the revenue with the producer.
This type of syndication model works fine for content that is evergreen, but there is simply no way to guarantee that a specific number of targeted viewers will watch the content
during a specific time frame. As a result, this type of syndication does not work well for viral videos, branded entertainment, and any other video assets that are campaign-driven.
Paid syndication has emerged over the past couple of years as a way of solving this problem by enabling audience guarantees through a media buy. Studios, brands, and their agencies are beginning to invest more heavily in paid syndication and the models employed to execute these campaigns have begun to evolve in response.
To date, the primary distributors of these types of video campaigns have been video ad networks. Networks provide the audience scale, the targeting, and ability to deliver a set number of views within a time window. Unfortunately, most of these networks only know how to deliver impressions and not video engagements.
In an effort to deliver a higher quality experience to brands, studios, and consumers, new networks that focus specifically on syndicating long-form video are emerging. To understand how these new networks differ from traditional video ad networks it's helpful to look at their place in the evolution of the paid syndication model:
In-Banner Autoplay: In-banner autoplay has been the default option for long-form video syndication for the past couple of years. The advantage to this type of distribution is that video views can be served one-to-one with page impressions and priced on a CPM basis. This has enabled ad networks to leverage their existing display inventory to deliver tens or even hundreds of millions of views over the course of a campaign at prices that sometimes drop below a penny per view. The problem with this approach is that a large number of impressions does not necessarily equate to a large number of engaged viewers (and often equates to a large number of irritated viewers).
Contextual In-Banner Autoplay: A few networks have begun to improve on the in-banner model by contextually targeting the video to the text in the page and providing consumption guarantees (such as minimum completion rates). Sometimes the video player is even delivered outside the ad gutter and positioned in-page (although still in a 300 x 250 size). The result is higher completion rates and improved engagements with the video. But again, the problem with this approach is that while it may minimize the number of wasted impressions it does not provide for a guaranteed number of engagements.
In-Banner Click2Play: More recently, a couple of networks have begun to offer guaranteed user-initiated views for in-banner media buys to solve the guaranteed engagement problem. These networks still deliver video in-banner like any other network, but they optimize against the CTR to guarantee a minimum number of users actually click to watch the video. Unlike autoplay in-banner units, in-banner Click2Play is priced on a per user-initiated-view basis or CPV. In-banner Click2Play represents a substantial leap forward in engagement, but it still results in the video being viewed as an advertisement rather than a piece of content.
In-Page Click2Play: Today, a new model is emerging that also guarantees engagements, but instead positions the video in-page as content rather than in-banner as an advertisement. These units are priced under the same CPV structure, but the video players are larger and they are promoted editorially often with accompanying articles. This type of placement feels more organic to the user and helps better capture the value of the investment in video content production.
This evolution has been in response to advertiser demand for increasing accountability towards the number of engaged viewers a network can deliver. Delivery of raw impressions for long-form video undervalues the content and treats the video like any other display advertisement. While the CPV pricing model is still foreign to some marketers, it is becoming a more familiar unit to more and more brands that focus on long-form video.As brand advertisers continue to gravitate towards delivery models that reward consumer attention and intent, we are sure to see continued evolution of the paid syndication model and survival of the fittest networks.
Thanks Alex, good info.
Is it possible for these Insider columns to become any more self serving? "Everyone does it wrong except for my company, which is the result of creating a more perfect network."
Any ad network can deliver impressions. End of the day, marketers must demand of all of their brand distribution partners that their end goal be met, whether that's clicks, video starts, video completions, conversions or some custom engagement and
can get here by paying for ad space only when those goals are met.
The ad networks that can price on engagement versus impressions will make syndication vs on-site placements a moot point
Hey Jason, I respectfully disagree. Remember, users are ignoring ads more frequently and in some cases being annoyed by them (eg. autoplay autosound). When big investments are made in content (studios, tv producers, branded entertaiment, etc), those content owners don't want to minimize the impact of that content by syndicating it as an ad unit (otherwise, why spend that much money on production, all you need is a rich media unit). Additionally, users are looking for real content that acts like content in the page, and the brand loves when a user expresses intent to watch that content and engage with it. Ad networks are fantastic delivery outlets but they will never match the effectively of true content delivered in-page.
So, even with all these evolving distribution form factors, what you get is limited to 4-5 metric - play, pause, completion, replay and may be 1 more. That is nowhere near the web form metric for click to action at different segments of content. While web page is spatially 4 directional, video is temporally uni-directional.. events on videos as timecode progresses. The magic happens if you are able to capture analytics on web style spatial clicks.. all over the contexts in a video. That's the new world.