An A/B test, also known as Split Testing and Bucket Testing, is a methodology in which two versions of marketing assets, like an ad message, imagery, or an app or a webpage, is tested proportionately across visitors to see which variants perform better. The test uses statistical data to evaluate your page and show which one drives a higher conversion rate.