Test A/B

What is an A/B test?

‍
The A/B test consists of developing and launching two versions of the same website or element within a page to measure and see which one works better in terms of conversion, sales or engagement with users.

When we do A/B tests, we create two random groups of users who visit our website. Group A is shown the original version and group B the modified version, which is the experiment we want to perform.

The experiments we conduct seek to generate versions of the website, or of the element we are testing, that generate a significant increase in some of our metrics: more sales, leads or micro conversions.

A/B tests are kept running until we have reached a data-based conclusion as to which version of the experiment works better: the original version (A) or the modified version (B).

This data-driven conclusion is called statistical significance and is determined based on the difference in conversions between the two variants and the sample size of users who have interacted with them.

Thanks to A/B testing, we can observe relevant trends in your users' behavior that will help you to detect improvements more efficiently than, for example, through market research.