A/B Test – How to run banner experiments on Axeptio
The A/B Test feature allows you to compare two consent banner configurations head-to-head to determine which one performs best in terms of consent rate. This is a powerful tool to make data-driven decisions about your banner design and wording.
⚠️ Requirements before you start
Enterprise plan only. This feature is exclusively available on Axeptio Enterprise plans.
Location-based display must be enabled. The A/B Test feature only works if location-based display is active on your project. Without it, the test cannot be launched.
Available for Brands and Publishers banners. You can run A/B tests on both banner types.
Configurations must be LIVE. Only configurations that are already live and active can be selected for an A/B test.
A configuration can only be part of one active test at a time. You cannot include the same configuration in multiple simultaneous A/B tests.
Multiple tests can run simultaneously. You can run several A/B tests at the same time, as long as each configuration is only involved in one test at a time.
Where to find the A/B Test feature
Once inside your project, navigate to the left-hand sidebar and click on A/B Test.
Creating a new A/B Test campaign
Click New A/B Test campaign to open the campaign setup page. The setup is divided into 3 steps.
Step 1 – Choose your country, regulation & language
Select the country, regulation (e.g. GDPR, CCPA), and language in which you want to run the test. This is important: you can only compare banners configured under the same country, regulation, and language combination.
Step 2 – Select 2 configurations
You must select exactly 2 existing banner configurations to run the test. Both configurations must:
Be currently live and active
Be of the same type (Brands or Publishers — you cannot mix the two)
Use the same language and regulation
Not already be part of another active A/B test
⚠️ Important: Once the A/B test is running, neither configuration can be edited. They will be locked for the entire duration of the test to ensure result integrity.
Step 3 – Set your test options
Configure the following options:
Test duration
Choose how long the test will run. The following durations are available:
1 week
2 weeks
1 month (recommended – allows you to collect a large enough sample for statistically meaningful results)
Custom period – define a specific start and end date
By default, both banners will be displayed to visitors in roughly equal proportions throughout the test period.
Get notified when it's done
Add one or more email addresses to receive a notification when the test has ended and results are ready to review.
Launching the test
Once all steps are completed, click the Launch test button. The test will start immediately and run for the chosen duration.
You can monitor results directly from the A/B Test section of your project dashboard.
Frequently asked questions
Can I edit my banners while the test is running?
No. Both configurations involved in the A/B test are locked for editing during the entire test period. This ensures that the comparison remains valid and results are not skewed by mid-test changes.
Can I run a test between a Brands banner and a Publishers banner?
No. A/B tests can only be run between banners of the same type (Brands vs. Brands, or Publishers vs. Publishers).
Can I run multiple A/B tests at the same time?
Yes. You can run several A/B tests simultaneously, as long as each individual configuration is only included in one active test at a time. A configuration that is already part of a running test cannot be added to another one.
What happens if location-based display is not enabled?
The A/B Test feature will not be available. You must activate location-based display in your project settings first.