September 2017
Beginner
306 pages
8h 13m
English
A/B testing is split testing between two versions of something--A and B. Half the people in your test use A and half use B. You measure the effect of the usage in each group and decide which version is better according to a metric. A/B testing is very popular as it is simple and fast to do, provided you have enough people. It is also great for showing which of two design decisions will have the most impact. However, it is also quite weak because you must only vary one thing at a time to be sure that it is the thing that is causing the difference between the groups.
In this way, A/B testing is an example of a controlled experiment. Everything except the test condition must remain as similar as possible in each ...
Read now
Unlock full access