A team runs an A/B test on a feature that only two hundred users see per week and declares a winner after four days. What is the flaw?

• AA/B testing needs enough traffic for the result to mean anything, and a small sample over a short window produces a difference that is indistinguishable from noise.
• BA/B tests are not compatible with Scrum.
• CThe test should have run for a full Sprint to align with the cadence.
• DA/B tests can only compare two versions, so a third variant was needed.