How do I analyze the results of an A/B test?

You've launched an A/B test and now want to see the first results? Here are more details on this topic.

Displaying statistics

In the Statistics tab of the Dashboard, you can get a visual overview of the results of your ongoing A/B tests at any time.

To do this, simply filter the statistics by selecting the context, the group, and then the scenario on which an A/B test is running, and enable the following field:

You will be able to observe a trend over a given period and quickly analyze the performance.

This can also help you determine whether the test is nearing its end or, on the contrary, whether it needs to continue.

Analyzing the numbers in the Dashboard

To correctly analyze your A/B test data, it is important to recall the points you will have defined beforehand:

Your initial hypothesis
A reminder of the two variants, the duration and the scope of the test
Your analysis KPIs

Once your test results are compared against your KPIs, you will be able to make the right decision and choose the best-performing version, or extend the test if needed.

It is important to take into account the distribution percentage between the two scenarios, especially if you have not split the segment equally between both scenarios (50% / 50%). It will therefore be necessary to properly weight the data for analysis.

Ideally, we recommend creating a test logbook to keep a record of the hypotheses tested (duration, scope, journeys, versions), as well as the recorded results.

And it doesn't stop there. Once a hypothesis is validated, test new ones!

Statistics in your own Analytics tool

To complement your analysis, if you have connected Poool with your own Analytics tool (GA4, Piano or others), you can find the performance attributed to your different versions.

Example of a Piano view:

💡 To clearly analyze your data, we recommend including "Version A" and "Version B" in the title of your scenarios. This way, you can easily find this information in the performance data reported in Piano or GA4.

You'll see, once you start A/B testing, you won't be able to stop...

If you have any questions, feel free to contact us via the chat!

Updated on: 20/03/2026

Was this article helpful?

Thank you!