• by Gabi Peles , July 18, 2014

As advertising budgets shift from TV to online, there has been a lot of excitement about what digital video can accomplish. With tools that enable powerful programmatic and personalized solutions, digital video advertising is generating high engagement, awareness and bottom-line results for brands. Seeing the power of this new generation of advertising is impressive.

But many advertisers aren’t truly seeing the results they could, and it’s not because customers aren’t seeing ads and making purchases. It’s simpler than that. They literally aren’t “seeing” results in digital video because the measurements they’re using weren’t designed to show the new data points that are now available. It’s impossible to identify true ROAS (return on advertising spend) without accurately connecting sales to the action that resulted in the sale, otherwise known as attribution. So why are brands and agencies still making mistakes in tracking how their marketing efforts are driving revenue? 

It comes down to a fundamental error in measuring attribution, one that has been inherited from advertising’s TV ancestry. If attribution looks at the actions that lead up to a given outcome, your relevant data points are limited in the TV world. TV as a medium of information consumption is passive, typically limited to consumption within the home. This passive nature means that advertising is primarily focused on reach and generating the highest number of eyeballs. And that’s what makes attribution break when calculating for TV advertising; despite efforts by Nielsen and others, it’s hard to connect general awareness achieved through TV to intent and ultimate purchase.

With the shift of dollars to online video, TV advertisers are trying to extend their reach measurement to measure ROAS. In doing this they are employing “last touch” methodologies from other digital advertising channels for calculating attribution rates, assuming that the last commercial a consumer saw was what led to a later action. This was a band-aid solution -- the only possibility given the nature of TV and the technology available for measuring the effectiveness of display video advertising. But last touch, or reach, is only the tip of the iceberg when we consider digital video, and it's irresponsible to ignore the range of data points available and use outdated attribution rates to measure modern digital advertising.

Brands, and ad serving platforms, must employ attribution models to accurately measure digital video. Digital video ads aren’t always designed to drive immediate post-impression action. Thus, in the same way that you can’t be sure a viewer’s exposure to a TV commercial was indeed the last touch they had with your brand, a viewer of online video may be simply watching video after video in a similar, passive “lean back” experience. If a viewer is watching but not clicking through, you must still properly measure impact with an accurate attribution model comparing exposed viewers to those who aren’t.

Not recognizing this connection in attribution is bad enough, but the wrongheaded thinking is even worse when we consider that it omits a huge range of other post-impression data that becomes available on digital. This data shows us the real-world ways consumers interact with and respond to online content, and can provide deep segmentation of reach, engagement and purchases, from search history to in-store interactions and more. In short, considering this data means we can know where our leads are coming from.

Brands, agencies and marketers still using last-touch attribution are diluting the true measurable value of their advertising in not taking into account the full, digital marketing pie, of which video is a major contributor.