How do you choose a layout for your online store? Which elements should go where and what sort of promotions work best? To find out the answers to these and other questions, you need A/B testing – a technique for comparing different versions of the same website. Also known as split or bucket testing, the A/B process reveals changes in the click and purchase behavior of website users when presented with different layouts, banners, promotions, etc. The following is an overview of how to make the most of A/B testing.

Test your versions simultaneously
In an A/B test, you run two versions of your website simultaneously. Some visitors to your site see version “A,” while others see version “B.” If you run the tests successively instead, the conditions are not identical and the results will not be reliable.

Use a good sample size
A/B testing with just 20 site users won’t tell you much about your store. Make sure your sample size is large enough to represent your customers in general and not just a small cross-section.

Don’t confuse your site users
Ensure that each test group always sees the same version of your site (“A” or “B”). If they see the versions changing, this may lead to confusion and they will leave the store.

Repeat your tests
Any A/B testing is a step in the right direction, but don’t be satisfied with just one test. Markets change quickly and your customers expect to see new promotions and a fresh design at regular intervals. Customer expectations are constantly evolving, so it’s important to be able to adapt your online store easily in response to test results.

Since your target audience is specific to you, you’ll achieve the highest conversion rate if you study the audience yourself. Different groups of customers respond differently under different sets of circumstances and can only be examined and adequately defined using A/B testing.