What does it mean to have a primary metric in an A/B test?

• A. It is the secondary measure used to validate the main outcome
• B. It is the main outcome used to judge the test's success
• C. It is not important
• D. It is determined after the test ends

Answer: (B)

In an A/B test, you choose one main outcome to judge whether the new variant performs better. This single measure—the primary metric—drives the decision about success. It’s defined before the test starts and is used to calculate how many participants you need and whether the observed difference is statistically meaningful. While other metrics can be tracked for context, they are not what determines the win or failure of the test. If your goal is to increase sales, the primary metric is the one that best captures that objective (for example, conversion rate or revenue per visitor) and is the basis for declaring the test a success.