Is Online Advertising Dead?

Posted on Posted in Web Development

During my time on the internet, as programmer and as a user, I repeatedly wondered how many people actually click the banner ads and google ads. After all, I am a proud AdBlock user and therefore rarely see any ads floating around. Even if it doesn’t catch all ads and one of those annoying pop ups (or overlays) appear, I close them even before they are propery loaded into the browser.

So, who are those people that make the online advertising branch so lucrative and profitable for Google, Facebook and Co? Well, I suspect it’s robots. A study shows that around 90 % of all clicks on ads on major websites are coming from robots [1]. This would make sense, after all. As real users do not click on ads anymore and become “banner blind”, companies resort to techniques that artificially increase the click numbers – and eventually their profits. The same report also states that there was not a single click that they were not charged for, concluding a prevalence of overcharging for ads for Google at around 95% and just marginally below that for Yahoo, LinkedIn and Facebook.

On the other hand, an eye tracking study also showed that people indeed do look at banners and remember those that are congruent with the content better than incongruent ones [2].

Testing Facebook

Let us have a look on Facebook. Veritasium, a YouTube science channel, did an interesting experiment and paid Facebook to “reach more people that matter to them”. And it worked almost immediately, accumulating thousands of likes. However, something was not quite right: the posts did not get any more engagement (likes, comments) than before, indicating that the new likes weren’t particularly interested in Veritasium itself. They made a video about it explaining the whole experiment in detail:

Case Study

As the owner and maintainer of different websites, I have also come across a very different domain, the analytics destruction domain. This domain is exclusively out to taint your analytics on your website, making so many requests that your results become useless and no measure for real users. Such domains can be blocked, but there are so many still out. As a case example, let me show you one analytics of one of the sites I operate. The site is situated in Austria, the content is solely relevant to Austrians and is in German. Clearly, there is no real incentive of visiting this site again and again if you do not speak German and are not Austrian. Well, let me show you the breakdown by country:

Country distribution of an Austria based website

This is not a special case, the distribution looks like this month after month. I have to add that Google Analytics shows a more positive picture here, with around 80 % clicks coming from Austria. Upon excluding a single domain referrer, 100 % came from Austria. I suspect Google Analytics automatically deletes a lot of known bots from the analytics.


With all that in mind, before you place your next order for any online advertising, make up your mind if you couldn’t somehow get to your real target group more efficiently, and maybe even save money doing that.

[1] Neal, A., Kouwenhoven, S., & SA, O. B. (2015). Quantifying Online Advertising Fraud: Ad-Click Bots vs Humans. Link to PDF

[2] Hervet, G., Guérard, K., Tremblay, S., & Chtourou, M. S. (2011). Is banner blindness genuine? Eye tracking internet text advertising. Applied cognitive psychology, 25(5), 708-716. Link to PDF

Leave a Reply

Your email address will not be published. Required fields are marked *