This article explains how A/B testing works and how to create an A/B test in Learnybox.
Use it when you want to compare two or more page versions, measure conversions, and automatically redirect visitors to the version that performs best.
A/B Testing: A Powerful Marketing Strategy
A/B testing is a powerful marketing strategy that compares two or more versions of a page to determine which one performs best based on a specific conversion goal, such as registrations, sales, or page visits.
During the test, you only need to share the link to one page. The system will automatically distribute visitors between the pages linked to the A/B test. Once a winning page is determined, all traffic will automatically be redirected to that page. The other pages in the test will no longer be accessible as long as the A/B test remains active.
Example: If Page C becomes the winning page, the links to Page A and Page B will automatically redirect visitors to Page C.
⚠️ To make the other pages accessible again, you must deactivate the A/B test.
How to Create an A/B Test
You can create an A/B test in two ways:
From the A/B Testing tool.
Directly from a sales funnel.
Option 1: Use the A/B Testing Tool
Go to Marketing → A/B Tests using the side navigation menu.
Click Create A/B test
When configuring your new A/B Test, you can choose between three conversion goals:
Total registrations – Compares the performance of at least two lead capture pages.
Total sales – Compares the performance of at least two sales or order pages with payment forms.
Page visits – Compares page redirections to measure which page attracts or redirects more traffic.
A default configuration is applied for each conversion goal. You can keep these preset values or adjust them from the advanced settings section before launching your A/B test.
You can configure the test with the following parameters:
Minimum visits per page: the minimum number of visits each page must receive before a winner can be determined.
Conversion rate threshold: the minimum difference between the conversion rates of two pages required to declare a winner.
Automatically determine the winning page before the A/B test ends: when enabled, the system can automatically calculate the probability of one page being the winner and select it before the end of the test.
When automatic winner selection is enabled, you can also define:
the minimum number of visits per page required.
the minimum conversion rate one page must reach.
the minimum probability required for one page to be considered the winner.
The minimum probability required for one page to be considered the winner represents a statistical confidence level (e.g., 90%) to ensure results are reliable and not due to chance. This parameter determines how sure the system must be before declaring a page as the winner. It represents the probability that the difference in conversion rates between pages is real and not due to random chance.
For example, if you set the confidence level to 90%, the system must be at least 90% certain that the better-performing page is not winning by accident. A higher confidence level makes the result more reliable but may require more data (visits and conversions) to reach a conclusion. This will help determine a winner when the performance gap between pages appears too quickly.
In simple terms: the higher the statistical confidence, the more trustworthy the result — but it might take longer to set a winner.
⚠️Reminder: Once a page becomes the winner, the other pages in the A/B test will automatically redirect to the winning page. To make the other pages accessible again, deactivate the A/B test.
Option 2: Create an A/B Test from a Sales Funnel
Within your sales funnel:
Select Create A/B Test for a sales funnel page (from the funnel page action menu)
Choose to compare the selected page with:
A duplicated version of the page, which you can then edit.
Or a completely different page.
When you choose to create the test from the current page, the system duplicates the page and places the copy just after the original page in the funnel. You can then edit the copy, including its copywriting, design, or other elements.
To edit an existing test, select Edit A/B Test from the funnel page action menu.
View A/B test results and manage tested pages
To view the results of an A/B test, go to Marketing → A/B Tests, then click the name of the A/B test you want to open.
From this page, you can track the performance of each page in the test, including visits, conversions, conversion rates, and the 30-day report.
You can also manage each tested page from the page action menu. For example, you can:
Deactivate page: temporarily disable a page in the test.
Set as winner: manually select a page as the winning version.
Reset page: delete the conversion metrics collected for that page and restart its tracking.
Delete page: remove a page from the A/B test.
⚠️Warning: Resetting a page deletes its existing conversion metrics, so use this option only if you want to restart the tracking for that page.
At this stage, you'll be able to set up your A/B test, and Learnybox can compare the selected pages based on your conversion goal.