What are common A/B testing mistakes to avoid?

Common mistakes include stopping tests too early before reaching statistical significance, testing too many variables at once, not having clear hypotheses, ignoring segment differences, changing tests mid-run, not accounting for external factors, testing low-traffic pages, and not following up on results. Other mistakes include testing elements users don't see, not considering mobile vs desktop differences, and implementing changes without proper analysis. The biggest mistake is not testing at all—even imperfect tests provide valuable learning. Example: A team stopped a test after 3 days thinking they had a winner, but after running it longer, the results reversed—they almost implemented a losing variant.