Contents of the A/B Test Configuration page

The A/B Test Configuration page provides settings to specify test splits and parameters that are required to send a mailing as an A/B test.

Configure A/B tests

Field Description
Percent of recipients for each test The size of each test split. For each split configured in the A/B test, a portion of the email recipients receives a different version of an email message. You specify the size of the recipient sample here, as a percentage of the total number of recipients in the list. The percentage is the same for all splits.
Percent of remaining recipients The portion of the recipient list that is not associated with a test split. Recipients that are not associated with a test split receive the winning version of the email.

Test splits (table)

Configure test splits by using the settings available in the table.

Circle with cross, Delete split Delete the test split

Up Down arrows: Change test order Change the run order of the test splits.

# The number of the test split.

Test Name: Enter a unique name to the split.

Document: Select a document that contains the layout, content, or subject line that you want to test. The list includes all published email communications that are not currently used by other mailings.

Multi-content: Click Edit to select a content element to test. The Configure Multi-content window displays thumbnail versions of the content that you can select if the document contains zones with multiple content elements. Elements that are controlled by personalization rules do not display. For each zone, select the content that you want to include in the email that is sent with test split.

Preview: Click to preview the version of the document that is sent with the test split.

Evaluate test results Specify how long the system waits after you start the A/B test before it evaluates recipient responses. Select a test duration (measured in hours) or select a specific date and time.

Winning criteria

Select the method that the system uses to evaluate recipient responses and determine the winner of the A/B test.

  • Maximum unique views The number of different recipients who opened the email.
  • Maximum unique clicks The number of different recipients who clicked at least one link in the email.
  • Minimum unique complaints The number of different recipients who reported the email as spam or asked to unsubscribe.