A/B testing
Comparing two versions
A/B testing is a method where you run two versions of the same page or element at once, version A and version B, and watch which achieves a better result, for example more orders. Visitors are split randomly between the two, so real data decides, not guesswork.
How A/B testing works
In an A/B test you create two variants: the original version A and a changed version B that differs in one thing, for example the button text, the headline, or the layout. Incoming traffic is split randomly between them and each group sees its version.
During the test you measure a chosen goal, most often the conversion rate, the share of visitors who took the desired action. Once you gather enough data, you compare which version did better and roll that one out for good. To keep the result trustworthy, only one element is changed at a time.
What A/B testing is for
A/B testing replaces guesswork with evidence. Instead of debating which headline is better, real customer behavior decides. Often even a small change brings noticeably more orders without having to increase traffic.
- Text and color of call-to-action buttons.
- Headlines and the main message on a landing page.
- Page layout and the order of sections.
- Form length and fields.
What to watch out for
A test needs to run long enough and on a large enough sample, otherwise chance skews it and the result misleads. The one-change rule applies too: if you alter several elements at once, you won’t know which one caused the difference. Testing is an ongoing process, not a one-off event.
Want to decide by data, not by guesswork?
Explore our marketing servicesFrequently asked questions
How much traffic do I need for an A/B test?
It depends on how large a difference you want to detect reliably. As a rule, the lower the traffic and the smaller the expected difference, the longer the test has to run so the result isn’t just down to chance.
What is worth testing first?
Elements that the most people see and that influence the decision the most: the main headline, the call-to-action button, and key elements on the landing page. That is where even a small change has the biggest impact.
What is the difference between an A/B test and a redesign?
An A/B test verifies a specific change against the original on real data. A redesign changes the site on a larger scale at once. After a redesign it therefore pays to fine-tune the key elements precisely through A/B testing.