Call us: +46 (0)8 410 413 13 Email us: support@ungapped.com
A/B testing of content in email
2021-10-08
With an A/B test of content, you can find out, for example, which headline, image or which button color gives the best effect.
Step 1. Enable A/B testing
- Open the Email Settings in the right menu
- Go to the Advanced -tab in the right menu
- Enable A/B-testing
- Now two subject lines appear above the email itself.
If you only want to A/B test the actual content of the email, enter the same subject line on both A and B.
(Our recommendation is to only A/B test the subject line, or to only A/B test the content. If the difference between A and B is too big, it is otherwise difficult to “put your finger on” exactly what led to the increased degree of opening/click.) - Select the percentage of your recipient list to use for the A/B test.
- Specify the duration of the A/B test (1-24 hours).
NOTE! If you have less than 100 recipients, your entire recipient list will be divided into two equal parts – half will receive the A-version and the other half will receive the B-version – and the e-mail will be sent at once. - Select what counts as “winner” – either the version that gets the most openings or the version that gets the most clicks.
Step 2. Mark which blocks in the email belong to the A-version and the B-version
- Mark a block to be A/B-tested.
- Go to the Advanced-tab in the right column
- Mark whether the block should belong to the A-version or the B-version.
(The blocks that should be visible in both the A- and B-version you don’t do anything with.) - Close
- If you open the preview (= the eye in the right column), you can preview the different version of the email.
- Save the mailing and Send out the email.
When the A/B test is complete and a winner is selected, the winning version is automatically sent to the remaining contacts on your recipient list.
If the A/B test ends in a draw, the version you marked with a heart will be sent out.
Example:
I have a total of 1000 contacts on my recipient list and have decided that 20% should be A/B-tested, and the version that gets the most clicks should be named the winner after 6 hours. I start the test at 12 o’clock.
10% (100 recipients) receive version A at 12 o’clock.
10% (100 recipients) receive version B at 12 o’clock.
80% (800 recipients) receive the winning variant at 6 p.m.