Results of Testing Adsense Blending

In a post about blending Adsense ads with your site, I showed how to blend the Adsense colors so they matched your site. This is largely considered to be more professional and Adsense ads that stand out.

A couple of readers called me on this, asking me if I’d done any split testing of blended ads versus non-blended ads. I hadn’t, but promised to do so.

This post has the results of that testing.

First, the methodology. I picked a page on one of my sites and tracked blended usage for a 14 day period. I then changed the colors on the ads in that page and tracked the usage for another 14 day period. The basic stats that I tracked were the average click through rate and the average earned per day of the period.

The non-blended ads used a background color that appeared elsewhere on the page, so the ads seemed to fit with the site, but still popped out from the white background of the text.

Unfortunately, the Adsense TOS restrict me from disclosing any form of click through rate specifics. Even aggregate data for the 14 day test periods is probably excluded from disclosure by the TOS.

So, I have to be more vague than I would like to be with the results.

One way of comparing the two results is looking at the relative values of each. The blended CTR was 2.3 times the non-blended CTR. To pick numbers totally at random, if the non-blended CTR was 2%, then the blended CTR would have been 6%.

The total earnings were about the same for the period, which has more to do with the cost-per-click I earned for the ads…this should not have been affected at all by the ad colors, so is more likely an external factor.

These results shouldn’t be considered conclusive, since every site is different and your visitors might react differently than mine. I also did not run the statistical analysis necessary to determine if the difference in CTR was significant, given the number of impressions. I’m happy enough with the practical difference between the two click through rates.

I’ve switched that site back to blended ads now. Hopefully the increased CPC will continue as the CTR gets back to normal.

2 Replies to “Results of Testing Adsense Blending”

  1. That makes sense. Years back I saw a study on web surfing habits, and most surfers don’t page down. They’ll read what’s visible, it won’t catch their interest, and they’ll look for somewhere else to click. An above the fold Adsense block is a natural for them to click on.

Leave a Reply