What is statistical significance in A/B testing?

Statistical significance in A/B testing indicates that observed differences between variants are likely real, not due to random chance. It's typically measured using p-values, with p < 0.05 meaning there's less than 5% chance the difference is random. However, statistical significance alone isn't enough—you also need practical significance (meaningful business impact) and sufficient sample size. Common mistakes include stopping tests too early, not accounting for multiple comparisons, and confusing statistical significance with practical importance. Use sample size calculators before starting tests to ensure you'll have enough data. Example: A test might show statistical significance with a 0.1% conversion difference, but that may not be worth implementing if it requires significant development effort.