Picking the right number of variations

You can test up to 127 different pages in an A/B experiment, but you can test far fewer variations and still get useful results. Keep in mind that the more pages you test, the longer it will take for your experiment to obtain significant results. It's best to consider which main changes to the page you want to try out, then whittle down your results from there.

To start, focus on creating a few variations with big differences. Then, after seeing which variations perform best, you can perform follow-up tests to narrow those variations down to a more specific design.

Was this article helpful?
How can we improve it?