👉 Configuring an A/B Test

📖 Definition

An A/B test is a type of feature that enables you to test the performances of a new version of an element on your website or application. 
After analyzing the results of your test, you need to decide which version has performed best according to the KPIs you wanted to reach (e.g., the conversion rate). You can then apply these changes directly to your website/application.

⚙️ Configuration

To configure an A/B Test, apply the following steps: 

1. From the dashboard, click Create a use case.
2. Select the AB Test template.
   
   
   [Basic information]
3. Fill in the name of your AB Test and its description.
4. Choose the primary and secondary KPIs you want to follow.
5. Click Save and continue.
   
   [Targeting]
   
6. Define the type of users who will see your feature. For more information, refer to the article about Targeting configuration.
7. Click Save and continue.
   
   [Variations]
   
8. Configure your variations by defining their flag's name, type (text, number, boolean, array, or object), and value.
   
   
9. Click Save and continue.
   
   [Allocation]
   
10. Define the percentage of the traffic you want to assign to each variation.
    You can choose between a manual or a dynamic allocation:
    The manual allocation enables you to determine the percentage to assign to each variation.
    The dynamic allocation automatically diverts new traffic to the best-performing variation depending on the results of each variation.
    
    [Overview]
    
11. Click Save and continue.
12. Check that every step has been configured correctly.
13. (Optional) Notify your teammates that the A/B Test is ready to go.

💡 Use case

Let's say that on your e-commerce website, you have noticed that your users have trouble proceeding to checkout once their order is complete. You think this may be due to the color of the button and the wording "Proceed to checkout", which would not be suited. You have come up with 2 solutions but don't know which one would bring the highest conversion rate.
In this case, you can create an A/B Test:

1. Create an AB Test use case on your Flagship account.
2. In the basic information step, select the KPI related to a click on the “Proceed to checkout” button. You need to have configured it in your codebase beforehand.
3. Select the conversion rate as a sub KPI type. 
4. In the targeting step, select “All users”.
5. In the variations step, configure the two variations with two flags of type Text as follows:
   For the first variation:
   
   For the second variation:
   
6. In the allocation step, assign 34% of your traffic to the original version, and 33% to the other 2 variations.
7. Check the overview of your AB Test.
8. Save and activate your use case from the dashboard.

Need additional information?

Submit your request at [e-mail]

Always happy to help!