An experiment with two variants, A and B
A/B testing is the process of comparing two variables, such as an ad creative or audience, to determine the most effective strategy. Usually you would have one variable of interest which you manipulate, and other variables are held constant.
For example, as part of your user acquisition strategy, you might want to test whether targeting Audience A outperforms Audience B for app installs via a particular ad. By A/B testing your ad to two different audiences, you will learn which audience leads to a higher conversion rate.