A/B test sample size calculator
How many visitors does your test need, and how many days will it take? Enter your numbers: the answer updates as you type.
Two-sided test of two proportions, with a Bonferroni correction when you compare more than two versions. Plan whole weeks, at least one.
What Verdicto adds to the calculator
Splits your visitors
A short snippet shows half of the visitors today's page and half the new idea, and records what they do.
Checks the data every day
Broken rows are removed and the split is checked, so you never act on a broken number.
A verdict, with its value
Ship it, roll back or keep running, how sure it is, and what the change is worth per year.
Questions
How many visitors do I need for an A/B test?
It depends on your current conversion rate and the smallest improvement you want to detect. A shop converting 3% that wants to detect a 10% relative lift needs roughly 50,000 visitors per variant at 95% confidence and 80% power. Use the calculator above for your own numbers.
What is the minimum detectable effect (MDE)?
The smallest relative improvement the test can reliably tell apart from luck. Smaller effects need many more visitors: halving the MDE needs about four times the traffic.
Why should a test run whole weeks?
Visitors behave differently on weekdays and weekends. Running full weeks, and at least one, keeps the comparison fair.
Can I stop the test as soon as it looks significant?
Not with a classic test: checking every day and stopping at the first good result turns luck into fake winners. Decide the length up front, or use sequential testing, which Verdicto supports, so results stay valid when you check daily.