AdSense Experiments Show How to Increase Website Revenue

The AdSense Experiments tool is an easy way to test ad optimization and potentially increase revenue.

Google offers the tool to website publishers who are eager to increase revenue with a minimum amount of effort. To find AdSense Experiments, click on Optimization in the left column and then Experiments.

The easiest way to get started is by turning on Auto Optimize. AdSense will run its own experiments to find the best ways of filling ad space. Otherwise, two types of experiments are available. They are Blocking Controls and Auto Ads.

Blocking Controls Experiment

The Blocking Controls experiment allows publishers to test blocking General Categories, Sensitive Categories or Ad Serving.

For example, a travel website might block the Beauty & Health Care general category because it isn’t relevant to a travel site. Likewise. A family-oriented website might block a sensitive category such as “References to Sex” — yes, there is such a category in AdSense. An Ad Serving test might block animated display ads because they slow server response time.

Auto Ads Experiment

An Auto Ads experiment can test the results of various ad formats such as in-page ads, anchor ads and vignette ads. It also can test ad load, which is the number of auto ads that AdSense displays on a page.

Promise Media ran an ad load test on a site owned by this company and found that a 50 percent increase in the number of auto ads led to a 384 percent increase in the expected revenue. Google automatically applied the change at the end of the experiment.

More Testing Tips

The various AdSense reports offer insights about what to test and what not to test.

For example, at least in the case of this company’s websites, the vignette ads do so well that they are not worth testing. Look for that report under Reports / Ad Behavior.

If a site uses mostly responsive ads, those reports also aren’t worth considering because the sizes change depending on the situation.

The Served Creative report may show that animated images have a low impression RPM. An AdSense experiment could potentially test blocking that category for the sake of a faster server response. But does it also mean fewer impressions to replace it? Will the replacement impressions have a lower RPM as well? An experiment may help answer those questions.

AdSense Experiments clearly is a useful tool in identifying ways of increasing website revenue with the help of Google. Site publishers should consider running tests on a regular basis to maximize their earnings with only a small amount of effort.