Using A/B Testing To Improve AdSense Revenue

Servando @ Stream SEO: I have been a lot more strict while receiving Guest posts in the last 3 months. From every 10 guest posts proposals I receive, I usually approve 1. Mostly because many of them are low quality posts with 500 words for the sake of a backlink. That’s why you’ve seen weeks where I just don’t publish anything even if I could accept guests posts. 

However, I find the A/B testing for AdSense articles very interesting and it really deeps into what I’ve been doing lately (I have an special posts coming around this too). Since many of you are niche bloggers, and AdSense/Amazon are key affiliates to make money, I’m happy to publish Ankit’s article.

Google AdSense is used by millions of website owners from across the world. While it was once a tedious task to find relevant advertisers for a website, AdSense was able to remove these barriers so that website owners could focus more on their website and less on finding advertisers.

Many AdSense Publishers are constantly looking for ways to increase their ad revenues and have never used A/B Testing. In fact, through the use of A/B testing, it is possible to generate higher RPMs (Revenue per thousand impressions) and do so with minimal effort on the part of the website owner.

A/B testing involves the testing of ads to ensure that the highest performance is always reached.

It helps you find the highest converting ad units on a website. Through A/B testing, the site owner will determine not only the behavior of their visitors, but also where they click more often and what entices the readers to click on ads more frequently.

Delving Into A/B Testing

A/B testing will provide site owners with statistics that relate to the return provided by Ad units. During testing, two or more layouts (different ad size/type/location) will be rotated among users. One user may see “A” and the next user may see “B.” The statistics gathered would show how one advertisement and placement performed over another. Perhaps users were more inclined to click on an advertisement that contained an image, or the use of a certain text color attracted more clicks. This change may yield a higher CTR and finding such units is the goal of any testing campaign.

improve adsense revenue design

Design, Colors and Blending

With testing, every last click and piece of data collected will relate to one core factor – user behavior. There are only so many variables that a website owner is able to control to ensure that their visitors will be inclined to click on an ad. The importance of ad design, color and blending will be of the utmost importance during testing.

Text Ads carry a certain theme. With AdSense, it is possible to control the design of a text advertisement completely. You also have control over the type of ad shown: image or text. Text ads allow for borders, fonts and colors to be changed to match the design of a website. Having complete control over these factors equates to different testing scenarios. Whether you believe it or not, small changes relating to the theme of text ads can really make a big difference in increasing your overall ad unit CTR.

A/B Test Strategies

A/B testing is not a new concept, which means that there are quite a few strategies that can be put into place immediately to increase revenue from AdSense. Through these strategies, it will be possible to learn which theme gets most clicks, what locations have a higher tendency to be clicked and what ad size provides the most clicks.

1. Ad Design

As touched upon previously, the design of an ad plays an important part in improving the Ad unit CTR. Different colors will have a very different outcome on the amount of clicks received. Ideally, during an A/B test, there will be a myriad of colors that are tested. The normal testing will include:

  • Warm colors
  • Cool colors
  • Blended colors
  • Contrasting colors

The testing of image versus text ads can also be done. While the Google team recommends advertisements that can contain both, many users have found success with one ad type over another.

improve adsense revenue

2. Location

You can place your ad units in different locations to see which location performs best. There are three main sections, which always need to be assessed with regards to location:

  • Above the fold: When an ad is placed in the top 600 pixels of a website, this would be considered above the fold. The idea is that every visitor to a site will view the advert no matter if they scroll down on a page or not. Oftentimes, ads that are above the fold perform very well.
  • In context: While a user is reading the content on a website, they are more likely to pay attention to any advertisements that are put next to or within the content they are reading. This can be done with text that is wrapped around an advertisement and melds right into the content itself.
  • After the content: Google has been an advocate of AdSense users placing an ad at the end of their content for years. The idea is that when a visitor is done reading through an article, they will want somewhere else to go. If an ad is placed at the bottom of this content, it is often clicked as a new place to learn more and further find answers to what the user is searching for.

3. Size

Advertisements come in many different sizes. Over the years, these ad sizes have changed dramatically to accommodate the use of mobile devices and larger screens. Larger adverts, such as skyscraper ads of 160×600 dimensions normally perform well on the sidebar of a website.

A 300×250 may perform better in these spots. Through split testing of ad sizes, a website will be able to determine which works best to increase their overall revenue.

4. Advertisement Amounts

Google AdSense allows publishers to place 3 ads per page. While some find it beneficial to use all of these ad slots, others find that they will make more profits when they only use one or two ad units. Different number of Ad Units on a page can also be A/B tested to see how it effects the Ad CTR and CPC.

Continuous Testing

A/B Testing can help you find the best ad unit variation for your website. However, there is one factor that can never be discounted – user’s behaviors will change. Once a website notices a significant decrease in what they deem a satisfactory CTR, it is time to start testing again. It is bound to happen. Sooner or later, your best performing ad unit will no longer be the same and the RPM will decline.

This is the reason why continuous testing and re-assessment should be done. This assessment will determine if:

  • Other ad sizes will perform better
  • Users have stopped clicking on ads and if recurring users have become “ad blind”
  • What ad locations are performing well and which are declining

There will always be room for improvement. We at AdPushup have doubled a tool, which automates the process of A/B Testing and Continuous optimization.

A/B testing allows a website to take advantage of their current users and ensure that they are earning as much revenue as possible.

Once a test has received enough data, it will be much easier to differentiate between which units perform well and which are not meeting their potential. A small tweak in color or placement can even lead to a 100 percent or more increase in ad revenue. As user’s behaviors change, it is up to the site owner to adequately adjust their advertisements and continue testing. This ensures that click-through-rates are always acceptable and that profits are always made.

About the Author: Ankit is a co-founder @ AdPushup, where he is building an A/B Testing & Optimization tool for online publishers. This tool will allow publishers to increase ad revenue on their websites from their existing traffic. Ankit is a programmer, online marketer, webmaster & growth hacker. He loves traveling and learning new languages.

 Email: We respect your email privacyEmail Marketing by AWeber 


  1. says

    Nice article indeed, I thank you for publishing the works of this guest author. I am only just getting into Adsense and this helps me know some important factors. Thanks!

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>