A/B Testing

A/B testing compares two versions of a page, email, or ad — with one variable changed between them — to see which performs better against a defined goal. Traffic is split between the versions, and the difference in outcomes is measured for statistical significance before declaring a winner.

A/B testing only produces trustworthy results with enough traffic to reach statistical significance — running a test on a low-traffic page and declaring a winner after a few dozen visits is a common, genuine mistake. This is part of why our AI Personalization & CRO category explicitly notes that these tools only make sense once you have enough volume for meaningful tests.

Central to every review in our AI Personalization & CRO category — VWO, Optimizely, and Unbounce's Smart Traffic feature are all built around this mechanic.