Once you’ve created and configured your A/B test, you’ll want to run it to begin gathering data on your Control Page and Variants. To run an A/B test,
Click the Review and Run Test button from the A/B Test menu.
Configure how your test should run. There are two configurations:
Traffic Split: the percentage of visitors that are randomly split between the Variants when visiting the Page. A visitor randomly assigned a Variant always sees the same Variant until the test is finished.
Confidence Level Required: the accuracy of the test results (i.e., when the winning Variant truly outperforms the other Variants). Typically you want to have the highest confidence level possible, but this impacts test duration. The higher the required confidence level, the longer it takes to declare a winning Variant. Choose the percentage based on your expectations.
The Estimated Time to Declare Winner field is also displayed. This is the estimated duration the test runs. This is calculated based on the selected traffic split, confidence level, and projected page traffic. Your page’s past traffic history is provided by Analytics Cloud.
Your A/B test is now running!
While an A/B test is running, you have two buttons available to help manage the test:
Terminate Test: terminates the test. To delete a test, you must terminate it first.
View Data in Analytics Cloud: redirects you to your A/B Testing dashboard hosted on Analytics Cloud. Here you can view your test’s traffic, reports, statistics, etc. related to your test. See the Monitoring A/B Test Results article for more information.
Awesome! You now have a running A/B test accumulating data based on user interactions with your Page. Next, you’ll learn how to monitor your A/B test’s results.