In an A/B test, which metrics are monitored?

• A. The chosen performance metrics, such as conversions
• B. Only page load time
• C. Only number of sessions
• D. Only bounce rate

Answer: (A)

In an A/B test, you keep your eye on the metrics you explicitly choose as success criteria for the experiment. Those chosen performance metrics are the ones you compare between the variants to determine if the change had the desired effect. Conversions are a common example, but any business-relevant outcome you care about—such as revenue per visitor, signup rate, time on site, or click-through rate—can be the primary metric. The key is that you define what success looks like up front and monitor those metrics to see if the new variant beats the control in a statistically reliable way. Focusing on a single metric like page load time or bounce rate only makes sense if that metric is the objective of the test; otherwise, you might miss the real impact of the change. That’s why monitoring the chosen performance metrics, such as conversions, is the correct approach.