The Drivers and Consequences of Ad Blocking: A Self-Filtering Mechanism That Increases Ad Effectiveness
By Evert de Haan. Published in the Journal of Interactive Marketing.
Give it to me in one sentence.
Ad blocking might not be as disastrous for brands and online publishers as first thought.
Give me a little more detail.
Marketing professor Evert de Haan devised a quasi-experiment* to find out whether or not it was worth it for publishers to try to combat ad blocking by disguising their ads or paying to be on white lists, and other things of that nature.
It was quite a convoluted experiment, with 10 different steps to ensure that participants couldn’t guess the true premise of the study, so we’ll just give you the barest outline of it here.
De Haan asked 600 participants to evaluate six news sites. Half of the participants saw the sites with ads, and the other half saw them without. During this phase of the experiment, De Haan asked the participants to use the mouse to indicate where they were looking on the site, as a substitute for eye-tracking.
Later on, the participants were then asked to name as many brands as they could within specific product categories, some of which matched the banner ads on the news sites. They were also quizzed about their attitudes towards the brands that appeared on the news sites (and of course, whether they used ad blockers in their daily lives).
The results showed that when ad-blocker users were shown news sites that contained ads, they spent less time reviewing them and rated them more negatively. People who did not use ad blockers however reviewed and rated the news sites more or less the same regardless of whether they contained ads or not.
De Haan also noticed a difference between ad-blocker users and non users when he asked participants to later list brands belonging to a specific category. As you’d expect, participants who were shown the news site with ads were more likely to mention the brands featured in those ads first when later asked to list brands within certain categories. But the difference was significantly less stark between ad-blocker users. The ad-blocker users who saw the ads were only 12% more likely than those who didn’t to mention the brands in those ads first in the recall test. But between the participants who didn’t use ad blockers in real life, the difference was 34.8%.
Essentially, says de Haan, that makes banner ads 190% more effective on non ad-blocker users.
*It was called a quasi-experiment because in real life ad blocking is a self-selecting behaviour rather than random. De Haan used something called propensity score matching to find participants who were all equally likely to use an ad blocker (but only some of whom actually did) to try to isolate the causal effects of ad blocker usage.
Why is this interesting?
De Haan’s conclusion is that forcing ads on people who use ad blockers is ineffective because they are more likely to ignore them or become dissatisfied with the website on which they’re hosted. If publishers only show ads to the visitors who are more likely to be influenced by them then marketers campaigns will be more effective, reasons de Haan. Does that work if brands are still charged for blocked impressions, though?
Any weaknesses?
De Haan mentions a few in the study, for instance that mouse tracking is not a perfect replacement for eye tracking, and that the brand mentioning test is not the most robust measure of ad effectiveness.
Where can I find the whole report?
Here, but it’s not free.
