You can use A/B testing, also known as split testing, to measure which variation of your page design performs the best. Half of your users will receive one version, and the other half, another one. This way, you can analyze which call to action, layout, colors or button styles get you more clicks or views, and choose the one that has better results for you.
The test has two main elements. The Subject is the element you are analyzing and the Goal is how you check if the design was successful. For instance, we can use a header design as the test subject and a CTA button as the goal. Any row, module or section can be assigned as a goal, even the test subject.
The Page Designer will present the results of your data using nice looking charts and graphs which you can even present to your clients. You can use this easy-to-use feature to run endless tests and keep on improving your website.
Start a Split Test
Go to the page settings and inside the content tab enable split testing. You can also right-click on the page and pick the split testing option.
You can define split testing variations.
There are additional options for the test you can choose:
- Bounce rate limit: define the number of seconds for a visitor to be considered a bounce. The bounce rate determines the number of users that land on your page and leave right away, so the lower the better. By default, a person that arrives on your page and exists within 5 seconds is a bounced visitor. You can adjust that limit here.
- Stats refresh interval: the stats with your results will update hourly, but if you want the reduce it to once per day.
- Shortcode tracking: you can enable this option to track off-page goals, which exist outside the tested page. Imagine you have a sales website with a process that ends on a “Thank you for your purchase” page when the client buys something. You can paste the tracking shortcode into the final page content and the page designer will track how many people completed the sale.
Choose the Test Subject
When you start the test, you have to decide what your test subject is going to be, by clicking over any element. When your test begins, the element will be duplicated and you configure the alternative version. You can make all the modifications you want, testing any number of changes.
After clicking the save button after selecting split testing to be activated, you will be presented with a YELLOW circle to choose the test subject.
Select the Goal
Now you need to select your goal, which can be any module, row or section. The page designer will now display different versions of your test subject to your visitors and track how often they click on the selected goal.
After selecting the test subject, you will receive a BLUE circle allowing you to choose the goal.
Successful Setup Confirmation
After selecting both the test subject and the goal, you should receive a confirmation asking you to configure the other possible variations of your test subject.
To scroll over the other variations available for your test subject, simply use the side arrows which appear when hovering the test subject.
Check Your Results
Your results will start to be collected and the page designer will show insights for each variation. You can view the results by clicking on the stats icon.
Several stats can be tracked:
- Clicks: how often the goal was clicked.
- Sales: how often a product was purchased.
- Conversions: how often a form was submitted.
- Bounce rate: how often visitors leave your page right away.
- Reads: how often visitors read the goal (for instance for a text goal).
- Goal engagement: this stat doesn’t consider the test subject, measuring how effective the goal is (ex: it’s not successful if it’s ignored).
End the Test
When you wish to end your test, right-click on the test subject and choose End Split Test or disable the test inside the page settings.
Once you end the test, you can pick the winner variation. The page designer will keep it and delete the other versions.