How to run a proper A/B test ?

Publishers, learn to run efficient A/B tests!

A guide for adops & yield managers in publisher monetization teams

Pursuing A/B tests allows you to proactively find ways to increase your revenue. It is recommended to A/B test any change you make to your ad stack to make sure you are going in the right direction. These can focus on timeouts, SSP incremental value, layout changes etc. However, running, setting them up, and understanding the data can sometimes be tricky.

Learn More

This guide will cover:

  • A list of the elements from your ad stack that you can a/b test
  • A step by step guide to implement a successful A/B test
  • Insights on how to analyze and understand the results of the test

More on this topic

We at Pubstack have been advocating for the extensive use of A/B testing in adtech for quite some time. In fact, this methodology was the only way to get viable and truthful metrics on many of our studies. You can find our whitepaper on monetisation that respects GDPR and user choice right here : "The guide to maximizing Monetization while respecting user choice"

Testing changes on your stack can only be verified through A/B testing, if you want to get your numbers (and your decisions!) right everytime. One of our customers has been using this method to measure the impact of each change to their ad stack, find out more by reading their Use case : Le Monde.

Other White Papers

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist us in our marketing efforts. See our Privacy Policy for more information.