Experimentation is at the heart of everything we do at Google — so much so that many of our products, including Analytics and AdSense, allow you to run your own experiments.
The AdSense Experiments page has allowed you to experiment with ad unit settings, and allowing and blocking ad categories to see how this affects your earnings. As of today, you can run more experiment types and have a better understanding of how they impact your earnings and users with some new updates.
Understand user impact with session metrics
Curious to know how the settings you experiment with impact your user experience? You can now see how long users spend on your site with a new “Ad session length” metric that has been added to the Experiments results page. Longer ad session lengths are usually a good indicator of a healthy user experience.
Ad balance experiments
Ad balance is a tool that allows you to reduce the number of ads shown by displaying only those ads that perform the best. You can now run experiments to see how different ad fill rates impact revenue and ad session lengths. Try it out and let us know what you think in the comments below!
Service announcement: We're auto-completing some experiments, and deleting experiments that are more than a year old.
To ensure you can focus your time efficiently on experiments, we'll soon be auto-completing the experiments for which no winner has been chosen after 30 days of being marked “Ready to complete”. You can manually choose a winner during those 30 days, or (if you’re happy for us to close the experiment) you don't need to do anything. Learn more about the status of experiments.
We’ll also be deleting experiments that were completed more than one year ago. Old experiments are rarely useful in the fast-moving world of the Internet and clutter the Experiments page with outdated information. If you wish to keep old experiments, you can download all existing data by using the “Download Data” button on the Experiments page.
We look forward to hearing your thoughts on these new features.