Prepare for a segment targeting experiment
Before creating a segment targeting experiment, it is important to be clear on what you want to achieve from it. There are some variables to consider in advance (based on your goals). These include:
Rail layouts
Target audience segments
Group allocation
Date range
Rail layouts
For your experiment, you may want to compare a new rail against an existing one, or create two new variant rails to see which performs best.
To set up new variant rails in the Rails Builder app:
Open the template that you want to use and click on the required layout. If you need to create a new template, see Creating, editing, and deleting templates.
In the Tools panel, click Add New Rail to create the first rail (variant A) to be used in the experiment.
Click Add Content… and add the required content items to variant rail A. When you have finished, click Save Unsaved Changes.
To create a second rail (variant B), click Duplicate rail from the 3-dot menu.
On rail B, click Manage Content to make the variations necessary for the experiment. For example, this could include adding content or changing its order.
View both rails together to ensure that they are correct.
Target audience segments
Audience segments are distinct sub-groups of users who have common characteristics, such as interests or devices used. Users are automatically assigned to segments based on their viewing behaviours. Grouping users in this way allows you to target your experiments more precisely.
Familiarity with the available audience segments will help you to make the best choices based on your goals.
Group allocation
In an experiment, groups are used to determine the rails that users can see. Each rail assigned to the experiment is allocated to either Group A or Group B.
Group allocation enables you to control the percentage of users in each group and therefore the percentage of users who can see each rail.
Date range
For any experiment, you should determine the time and duration that will best support your goals.