What is A/B testing / split testing?
An A/B test, also known as a split test, is a method through which you test two or more different versions of a webpage to gather data on which one performs best with your target audience.
In practice, a control group is exposed to the original version of the page, while another group — the variant group — is presented with a different version. Traffic is often split 50/50, so that 50% see the original version while the remaining 50% see the variant.
The test is then left to run while it collects data on visitor behaviour across the two versions of the page. Once enough data has been collected, statistical models and analyses can be used to determine which of the two variants performs best for a given conversion goal.
Depending on the test results, the variant is either implemented on your website or left out if the original has performed better.
You can also run A/B tests within your advertising on platforms such as Facebook and Google Ads. Here, a control group is likewise exposed to one version of the ad while another group sees the variant.
Why should you run A/B tests?
By running A/B split tests on new initiatives on your website, you ensure that you make data-driven decisions about changes to your site. Far too often, decisions can end up being based on gut feeling and subjective judgement about what looks best and should be implemented.
But ultimately, it is your target audience and visitors who "decide" what performs best — and therefore what is best for your business. This is precisely what A/B tests can help you find out.
Concrete example of an A/B test
Below is a concrete example of an A/B test we have set up.
In this test, we wanted to investigate the hypothesis that the search field on the website in question should be made larger, in order to encourage more visitors to use the site's excellent search function — making it easier for them to find their desired product and thereby ultimately increasing the conversion rate.
Control (Original)
Variant
How do you run A/B tests?
In practice, A/B tests are set up using third-party tools that are integrated into your website — typically via a script in the website's <head> section.
Some of the most well-known and popular tools for A/B testing are listed below:
- Google Optimize (Free)
- VWO (Paid)
- Optimizely (Paid)
- AB Tasty (Paid)
Ready to optimise your website or ads with A/B testing?
Call us on 30 12 42 72 or fill in the form below for a no-obligation chat about how A/B tests can help you improve your conversion results!
