Statistical Power
The probability that a test will correctly detect a real effect.
Power (usually aimed for 80% or higher) is the probability that your A/B test will find a statistically significant difference between variations if a real difference of a certain size (the MDE) actually exists. Low power increases the risk of Type II errors (false negatives) – missing a real winner. Power depends on sample size, baseline conversion rate, and MDE.
Ready to Run Better Tests?
Put this knowledge into practice! Test It is the easiest and most affordable way to run A/B tests. Get started for free.
Try Test It Now