A/B test duration calculator
Enter your traffic, baseline rate, and the lift you want to detect to see how many days the test should run.
Test parameters
Traffic that will be split across the variants each month.
Smallest relative lift you want to detect.
Traditional A/B testing
51 days
Fixed-horizon frequentist test: read it once the full sample lands, at 95% confidence and 80% power.
With Spar
16 days
Spar checks the result continuously and declares a winner once chance to win passes 95%, about 69% sooner on average when the true lift is at least your MDE.
Visitors per variant
50,757
Total visitors needed
101,514
How test duration is estimated
Duration is just the sample size your test needs divided by the traffic you can send it each day. The calculator first works out how many visitors per variant are required to detect your target lift, multiplies by the number of variants, then divides by your daily visitors to get the number of days.
Because the sample size depends on your baseline rate and the lift you want to detect, low-traffic stores and small target lifts produce long durations. Daily visitors are derived from your monthly figure, so enter the traffic that will actually be split into the test.
Use duration as a go or no-go check
The most useful thing this calculator does is tell you before you start whether a test is even feasible. If detecting a meaningful lift would take six months, that is your cue to test a bolder change, pick a higher-traffic page, or accept a larger detectable effect. Decide the duration up front and hold to it, since stopping early the moment a result looks significant is the fastest way to ship a false positive.
Frequently asked questions
Stop guessing what to test
A calculator tells you whether a test is worth running. Spar finds the tests worth running: it audits your storefront, ranks the gaps by revenue impact, and drafts the variations. Start a 14-day free trial.