What is “view-through” and how do we know that it’s real?
It’s possible you’ve never heard of the term “view-through attribution,” if that’s true, here’s the idea in a nutshell.
Visible ad impressions have value – the simple act of putting an ad in front of a human may influence that human to interact with your brand when they may not have, otherwise. That’s where the “view-through” part comes from – someone may have seen your ad and, even though they didn’t click or tap on it, that ad might have inspired them to search for your brand on Google or even to type your web site address into their browser. The “attribution” piece refers to the scientific effort to understand exactly which interactions may have been influenced by a viewed ad, and how much value should be attributed to that view.
View-through attribution gets a lot of coverage in digital media circles, but the underlying mechanic has been around as long as advertising itself. Aside from the rather unfortunate advent of the :cuecat and similar gimmicks, no one has ever clicked on a magazine or newspaper ad – let alone a TV commercial. Even so, it’s not uncommon for marketers and executives to proclaim they “don’t believe in view-through.” In some ways, I don’t blame them. SEM (search engine marketing) has been a massive digital channel for many years now and it operates almost entirely on a click-through attribution model. Someone searches for your product, they click on your search ad, they buy your product – you can connect all the dots and see exactly where, and how much, value was created for your business. Who wouldn’t love that kind of insight and simplicity?
Display media types – videos, banners, native ads, etc. – can also, generally, be purchased on a “per click” basis, but very few humans actually click on digital ads, making it much less clear to understand the ultimate value of those ad impressions to your business. In a (now very old) study, ComScore estimated that about 85% of display ad clicks came from just 8% of the internet population. And it stands to reason that percentage has decreased over time. That’s not to say that humans don’t respond to display media ads, they just don’t necessarily poke them.
As previously noted, there is a lot of great material available out there on the ins and outs of different approaches to view-through attribution – hit up your favorite search engine, it can make for some fascinating reading. But in this post, I want to share a recent client launch that essentially demonstrates the existence of view-through activity. This client never promoted their brand or web site prior to the launch of our native and SEM campaigns – as such, we had the opportunity for a great “before and after” look at traffic to their site. Here’s a chart from Data Studio, connected to Google Analytics, showing site traffic before the launch and after:
Nothing particularly unusual or insightful here, other than to note the fact that running promotional media was an effective way of bringing new users to the site. But then if we look at the traffic source data for the same time period, we see something a little more interesting:
Neat, right? The site was live on 1/12, but promotional campaigns launched on 1/14. We can see the expected traffic from Display (native) and Paid Search (SEM) starting on 1/14. But what we also see in virtual lock step with those promotional sources is Direct traffic! That means people are typing in the web site address to their browser address bar and visiting the site directly. Unless you want to try and make the argument that people spontaneously became aware of a new web site, including the URL, the obvious conclusion is that, in this case, Direct traffic to the site is generated by views of promotional media impressions. And it’s not immaterial volume - Direct represents about 17% of total traffic to the site for the period of 1/12-1/31, that’s second only to Display.
So, there you have it folks - actual evidence of view-through response in action. Please don’t take this as a resounding call to go plaster the internet with display ads in the hopes of getting “all the view-through” – that’s not what we’re saying. You should be thoughtful in your strategy, methodical in your approach and rigorous in your measurement of display media campaigns. We’d simply encourage you to give some consideration to where and how view-through activity and attribution fit into that effort.
If you’re wrestling with your own display media strategy or approach – drop us a line, we’d be happy to chat. Thanks!