AdSense Split Testing

Posted by Motti 1 July, 2008

We all want to know which ad give us the results. We know we can change the colors of the ad elements, the type and even the location it’s place in out website.

But sometimes we want to check what improve our revenues and which one not.

One famous trick that is used by many advertisers is called split testing or A/B testing.

The theory behind split testing is putting two separate ads in the same location with only minor change between them, testing both ads on the same time and location can help in finding which ads give us the best result.

Google don’t give the right tools to perform this kind of test but with the help a very simple JavaScript script we can accomplish this task very quickly.

Here are the steps to perform a simple split testing for two AdSense ads.

Create 2 separate custom channels, one channel for each ad.
Create 2 AdSense blocks with one minor change between them, for example different link color, use the channels you create in step one for those ads, different channel for each ads.
Located the place where you want to the put your AdSense block inside your website source code, and place the code below:

   1:  <script type=”text/javascript”>

   2:     var random_number = Math.random();

   3:     if (random_number < .5){

   4:         // Put here the first Adsense ad code

   5:        } else {

   6:         // Put here the second Adsense ad code

   7:     }

   8:  </script>

The code above will split the numbers appearances between the two AdSense ads, both versions will be displayed the same number of times.

After few a week you can track your AdSense revenues for each of the above ads using the channels that you used. You can find out which ad format version is more profitable.

If you want you can take this test even further and try to improve the better ad version again until you will find the best AdSense version for your website.

If you take the next steps remember to create two new channels for the two new versions.

Learn to find out which ads are more profitable to you and delete the other ones, after all it’s all about tracking.

Bookmark It