Manage Test Groups

This procedure walks you through the steps needed to create, edit, or delete a test group in Audience Lab.

Create Segment Test Groups

Prerequisites

• You need to have at least one conversion trait set up. You can set up conversion traits in the Trait Builder, by selecting conversion as the event type. For more information on what conversion traits are and how to set them up, we prepared a video for you.

  IMPORTANT

  Folder traits are not supported by Audience Lab. Setting the Event Type of a folder trait to conversion will not generate any data in Audience Lab for that specific folder trait.

• For companies using Role-Based Access Control: Assign the Audience Lab wildcard permission to User Groups to provide access. This permission allows the user to create and view the results of a test. A user will only be able to use segments from a data source they have read and map to destination privileges for. The user will only be able to use conversion traits from a data source for which they have “read” permissions. A user will only be able to see destinations they have access to as well. So, before adding the Audience Lab wildcard permission to a group, make sure the group has:

  • access to read relevant conversion traits;
  • access to read and map relevant segments for tests;
  • access to relevant destinations.

To create a new Segment Test Group:

1. Select Create New Test Group in the Audience Lab dashboard to start the wizard.

2. Basic Info & Choose Segment

   • Fill in a Test Group Name and a Description.
   • Choose the Base Segment either by navigating in the file browser or by typing in the search bar, confirm by pressing Choose Segment.
   • You can save the test group as a draft and resume working on it later.
   • An alert will show up in case the base segment you selected is already used in other test groups. Using the base segment twice may distort the conversion results for both tests.

3. Allocate Test Segments

   • You can create up to 15 test segments and divide the percentage of devices as you wish.
   • You can edit the name of the test segments by clicking on them.
   • The percentages automatically divide evenly to 100% when new test segments are allocated. You can then manually edit the percentages. Click the checkbox after editing the percentages and make sure they add up to 100%, otherwise you will not be able to proceed to the next step.

4. Set a Control Segment

   • Select a control segment if you want to set aside a certain part of the segment to be used as a control group. Control groups allow you to see the impact of the test segments you created compared to a benchmark.
   • You can select a test segment as control segment in the drop-down list, or you can choose None for no control.
   • Click Next when you’re done.

5. Select Conversion Traits

   • Add conversion traits by typing in the conversion trait window. This is a mandatory step and you cannot proceed to the next step unless you add at least one conversion trait.
   • You can add up to 5 conversion traits if you wish.
   • An alert will show up in case you select a conversion trait already used for other test groups.
   • Note that Audience Manager does not support using folder traits as conversion traits. Selecting a folder trait as conversion trait results in 0 aggregate and trend reporting displayed within the test.

6. Choose Destinations & Dates

   • Type in the desired destinations in the search field or use the drop-down arrow. Audience Lab test segments can be sent to URL, cookie, or server-to-server destinations.
   • Drag & drop segments to destinations.
   • After dropping a segment in a destination, fill in the Destination Mapping Value in the blind.
   • You can send the same test segment to multiple destinations and you can add multiple test segments to a single destination.
   • Destinations are grayed out if they are not available for a certain test segment based on Data Export Controls.
   • Users will only see the destinations they have access to based on the RBAC User Group they belong to.
   • Finally, you are required to select a start date for your test group. This date marks the start of the period in which your test group will be published to destinations. Select No End Date for an indefinite comparison of the test segments.
   NOTE

   Profile Merge Rules with an authenticated profile are only supported in Real-time destinations. If a test segment with a profile merge rule of that configuration is sent to a file-based server-to-server destination, the audiences might not populate.

   Click Next to review and finalize your test group.

7. Summary & Finalize

   • Review the information you added in the previous steps and select Finalize Group.
   • Remember that once you finalize a test group, it can be duplicated or deleted, but not edited.
   NOTE

   • You can save the test groups at any point in the creation process and return to the wizard at a later time. The test group status will be Draft and the test group will not send any data to destinations until you finalize the segment test group.
   • For draft tests, you can go back and edit the test groups by clicking Edit in the test group card in the main Audience Lab view.

Edit Segment Test Groups

In Audience Lab, you are only able to edit draft test groups. In the Create Segment Test Group wizard, you can save your test group as a draft and resume working on it later.

1. Navigate to the Audience Lab main view.
2. Search for your draft test groups and select the Edit control in the test group card.
3. Resume the Create Segment Test Group wizard and select Finalize Group when you’re done.

Delete Segment Test Groups

1. Navigate to the Audience Lab main view.

2. Find the test group you want to delete. You can either:

   • press the Delete control in the test group card, or
   • press the test group title in the test group card to go to the Test Group Information view and press the Delete control in the title bar.

3. For completed test segments, an alert will prompt you to download the CSV file to save the reporting if you wish.