There’s been a lot of bad blood about display ads flowing through the marketing industry of late. Some are misguidedly dismissing display advertising’s relevance all together, and even the banner ad’s creators are getting their punches in. Display advertising’s opponents love to point to a study by comScore and Starcom MediaVest that found that 84% percent of internet users never click on an ad, and also that the clicks from the remaining 16% don’t correlate to actual on-site conversions.
No ad to show here.
Those stats are accurate, but they’re not proof that display advertising doesn’t work. Rather, they’re proof that we’re measuring display advertising in the wrong ways. Years ago, digital marketing got obsessed with the idea that clicks would be the holy grail of marketing measurement, even though that view ignores the fundamental way we’ve gone about advertising for all of civilization: we give people something to see, read, or watch, and then we track whether they subsequently make a purchase.
In digital marketing, measuring a display ad’s success in that fundamental way is known as view-through attribution. When you’re served an ad, it often drops a cookie. Later on, if you visit that advertiser’s site, the cookie lets the advertiser know that you previously viewed one of its display ads. This is known as a view-through visit, or post-impression visit, and the ad that led to the visit is the view-through impression.
If we agree that clicks are a pretty much useless metric, then view-through attribution has to be the way that we measure display advertising. After all, an unclicked display ad works the same way as offline ads in magazines, billboards or buses. The difference is that offline advertisers don’t know which customers previously saw one of their offline advertisements. With view-through, you know which customers saw your ad, which will allow you to create much more effective campaigns in the future.
The crucial detail is the view-through window — the length of time that an ad impression receives credit for driving a view-through conversion. When marketers first started experimenting with view-through, they usually set a default window of 30 days — which is far too long for most campaigns. If you make a purchase on Amazon, chances are it isn’t due to a display ad you saw 29 days ago. Seven days is a much more logical view-through window, and the windows can be set for as short a period as 24 hours.
View-through’s reputation is still battered from some fraud during its early days of implementation, when scammer ad firms would buy up millions of super cheap impressions so they could take credit for almost every conversion on a site running a display campaign. But this scheme, called “cookie stuffing,” has since been eradicated, view-through attribution is now a reliable and important metric. Without it, we’ll never know display advertising’s true impact.
