Setting up an A/B test with two subjects
Laposta does not have a built-in feature for A/B testing, but you can set up a test yourself using an extra field and segments. This allows you to send two versions of the same campaign to a small portion of your list, for example to see which subject line yields the highest open rate. You can then send the best-performing version to the remaining subscribers.
Things to consider with an A/B test
A test is only truly reliable with a large list, though you can of course always test whether you see major differences in the results. Because the sending time can also cause a difference in the open rate, it is best to test one thing at a time: for example a different subject line, or a differently worded or designed call to action (the link to a relevant action in your newsletter).
Setting up an A/B test yourself
In this example, you send two groups (A and B), each consisting of 25% of your subscribers, a different subject line. The remaining 50% then receives the winning version.
Divide your list into test groups
Select a representative part of your list to receive version A or B.
- Export your list to Excel and open the file. If you want to import a new list for this test, open that Excel file instead.
- Add a ‘Version’ column and enter an
Aor aBfor a relevant percentage of subscribers. Leave this field empty for the remaining subscribers. - Save your modified list and import this file into your account. When importing, create a new multiple-choice field called ‘Version’. If needed, read Importing a large subscriber file.
Make sure at least 10% of your subscriber list has an A in this field and an equal percentage has a B.
Create segments for version A and B
- Go to your list and click on ‘Segment’.
- Click on ‘Create new segment’ and choose ‘subscriber data’.
- Select the ‘Version’ field and specify that it must equal
A. Save this segment asVersion A. - In the same way, create a segment
Version Bfor version B.
Create two campaigns and schedule them simultaneously
- Create a campaign. In step 1, indicate in the working name (for internal use) that this is version A. In step 2, select only the
Version Asegment under the recipients. - Do not send the campaign yet once it is ready, but return to the overview of all campaigns.
- Make a copy of version A. In step 1, change the working name to version B and change the subject line. In step 2, select only the
Version Bsegment under the recipients. - Once you are satisfied with both campaigns, schedule them in step 6 (‘Schedule delivery’). Preferably choose the exact same date and time, so that the delivery time does not affect the test.
Determine the winner
Wait a representative period for the results, and at least two hours, so that the subscribers in both test groups have time to open your newsletter. Next, create two segments to see which version was opened most often.
- Go to your list and click on ‘Segment’.
- Click on ‘Create new segment’ and choose ‘campaign results’.
- Select the subscribers with the condition ‘b.6 - have opened a specific campaign’ and choose the campaign for version A.
- Save this segment as
has opened A. - In the same way, create the segment
has opened Bfor version B. - On the ‘Segment’ page, check which segment is larger. If the
has opened Agroup is larger thanhas opened B, version A was opened more often and has won.
Send the winning version to the remaining subscribers
- Create a segment for the subscribers who have not yet received version A or B: go to your list, click on Segment → ‘Create new segment’ and choose ‘subscriber data’. Select the ‘Version’ field with the option ‘is empty’ and save this segment as
Remaining recipients. - Go to the ‘Campaigns’ page and copy the winning version. In step 2, select only the
Remaining recipientssegment under the recipients and send the campaign immediately in step 6.