The best way to optimize performance is to continuously test so you can weed out the weaker messages. You can now run an A/B test for any Broadcast copy and/or image! This helps you learn which messages are converting, without spending hours testing manually.

How to Run an A/B Test

First, head to Broadcasts tab. You'll see a new option towards the bottom of the Broadcast Wizard to “Create A/B test."

Step 1: Message

  • Here, you'll create two messages to test which version performs best.

  • We recommended that you select your Audience and write Variant A first. This will save you time copying and pasting for Variant B, as Emotive will automatically create a duplicate of the A version.

  • Clicking the left arrow lets you expand the variant.


  • Once the window is expanded, you can edit your copy or image.

  • Any changes you make to your message will update automatically in the right preview window.


Note: You cannot have identical variants (i.e., the same message AND same image) to proceed with creating your A/B test. Emotive will show a red error message if this is the case.


Step 2: Test Settings

  • Selecting the A/B test size determines the portion of the customer segment that will be split to test.

  • Selecting a test size of 100% will split segment in half, equally between the two variants.

  • If a test size is less than 100%, a winner will be determined and the winning variant will be sent to the remainder of the segment.


  • Next, choose how the winning variant is determined based on:

    • Highest click through rate

    • Highest conversion rate

    • Highest response rate

    • Highest revenue

    • Lowest opt out rate

  • You can also select a timeframe for the winner to be chosen.

Step 3: Preview

  • Finally, you can preview both variants and send your self a test before sending.

  • Once everything looks good, click "Review and send" and you can schedule or send the test right away!


Step 4: Review and Send

  • Scheduling an A/B test works the same way as a regular Broadcast. Whichever time your choose will be when the A/B test starts.

  • You can see the send time for the winning variant at the bottom. (The same time the winner is chosen by the system)


  • Use the review and send modal to give your copy and content one last review before sending or scheduling your message

  • You can use the "Back" option to return to editing your content


Have any feedback on A/B testing? We'd love to hear from you! Submit a request on Canny to let us know how we can improve this feature.