Smart A/B Test


Smart A/B Tests allow you to find the best-performing text element on any landing page easily with fully-automated AI-powered A/B tests.

In this artice, we’ll show you

How do Smart A/B Tests work?

Our Smart A/B test automatically cycles through all your variants, one by one. It starts by finding a winner in the initial test, then pits the next variant against this winner, continuing until it identifies the top-performing option.

This means once a test concludes with a clear winner, the system seamlessly proceeds to the next comparison, saving you time by handling all the setup and decision-making automatically.

What to test with Smart A/B tests?

OptiMonk’s Smart A/B Tests let you optimize all text content on your landing pages, including:

  • Headlines and subheadlines
  • CTA (Call-to-Action) button texts
  • Product and service descriptions

How to setup a Smart A/B test

  1. Log in to you OptiMonk account at
  2. Click on New campaign from the Dashboard, Templates or Campaigns menu.

new campaign.png

3. Choose the Optimize a website campaign type.


4. Then choose Smart A/B Test under Optimize from the list.

SAB_optimize (1).png

5. You'll see a modal displaying all the essential information about Smart A/B Testing. Once you're ready, click on Get Started.


6. Select or add the domain you’d like to run your Smart A/B Test on.


7. Next, you'll have the option to let OptiMonk recommend landing page elements for optimization (Get recommendations), or you can skip this and directly create your own A/B test. If you choose to skip, proceed to Step #10.


8. OptiMonk AI will automatically identify headlines that could benefit from Smart A/B Testing. Select any headline to view AI’s recommendations.

SAB_reco.png9. Once you select a headline, OptiMonk will display variant recommendations on the left, which can be A/B tested. You can edit and modify these by clicking on Edit and Preview.


10. OptiMonk will launch your website within its Dynamic Content Editor. To start the A/B testing process, navigate to the specific page you wish to test using Navigation mode. Once there, switch to Edit mode to begin making changes to your page. Click on the element - for example the headline - that you’d like to test.


11. The smart A/B test will automatically generate alternative headlines for you to test against your original headline. Feel free to modify or delete these as needed.



12. By clicking on Preview, you can view how your page appears with different headline variants. You can copy the URL of the preview to share with your teammates or open the preview to see how each variant looks. Use the arrow in Preview mode to cycle through and view each headline variant.


13. After adding your variants and clicking Save, you will be prompted to specify the pages where your A/B test should run. For instance, if you're conducting the A/B test on your homepage, you would set the URL rule accordingly:


14. On your Campaign Summary page, you'll find a list of all the elements you're A/B testing, along with all the variants for each element. The default conversion metric is set to visitors with orders, but you can change this to any Default or Custom Conversion Event.


How to track a Smart A/B test’s results

Starting the test

On your campaign details page, you can view results for each A/B test, including every variant tested. After activating the test, a message will confirm that your test is running and data collection is underway.


Running tests

Once sufficient data has been collected, at the top of the page, OptiMonk will provide an estimated uplift that you could achieve by implementing the winning variants of each A/B tested element. (= the highest uplift from all your test rounds)

For each element tested, you'll see a breakdown of all test rounds. OptiMonk automatically highlights the winning variant compared to your original variant, showing the conversion uplift and the probability of winning (=Chance to win).

Campaign (4).png

Concluded tests

When does a test end?

A Smart A/B test concludes under one of two scenarios:

  1. Clear winner identified: The test ends once a variant shows significantly higher conversions than others, and after it has run for several days to cover various traffic periods.
  2. Inconclusive results: If no definitive winner is determined after three weeks, the test round will end, acknowledging that the results are inconclusive.


What happens when a test ends?

When a Smart A/B test concludes, OptiMonk continues to display the winning variant of the previous round. If no clear winner is identified, OptiMonk will revert to displaying the original variant.

✉️ If you need any further help or have any questions, please let us know at, and we would be happy to assist you :)

Have more questions? Submit a request