A/B Test Sample Size Calculator — Free, Two-Proportion Z-Test
Free A/B test sample-size calculator using the two-proportion z-test. Set baseline CVR, MDE, and confidence level — get the per-variant sample size needed.
A/B Test Sample Size Calculator
Per variant
31,196
Total (A + B)
62,392
Target new rate
5.50%
Assumes 80% statistical power. Sample size scales with roughly the square of the effect you want to detect, so halving the minimum detectable effect costs about four times the traffic. Detecting a 10% lift on a 5% baseline needs about 31,000 visitors per variant; asking the same test to spot a 5% lift on a 2% baseline needs about 315,000.
You're on 7BusyBoss — 300+ free tools that run instantly in your browser. No signup, nothing uploaded.
Work out the sample size for an A/B test
Enter your baseline conversion rate, the minimum detectable effect (MDE) you care about, and your confidence level — get the number of visitors needed per variant before the result is trustworthy. It uses the two-proportion z-test that powers most experimentation platforms, including Adobe Target, Optimizely and VWO.
The inputs that matter
- Baseline CVR — your current conversion rate.
- MDE — the smallest lift worth detecting (a smaller MDE needs a much bigger sample).
- Confidence & power — usually 95% confidence and 80% power.
Why sample size matters
Calling a test early is the most common A/B testing mistake — early "winners" often regress to the mean. Fix your sample size up front, run until you hit it, and only then read the result. Always-on programmes can use sequential testing to peek safely, but a fixed-horizon sample size is the reliable default. Everything is computed in your browser.
Work out the sample size before the test rather than after it. Once results are in, the conversion rate calculator turns them into a rate and a relative lift — and reminds you which of those two numbers you are quoting.
How to use the A/B Test Sample Size Calculator
Takes about a minute. No signup, no download, your data stays in your browser.
- 1Open the tool. Scroll up to the A/B Test Sample Size Calculator above — it loads instantly in your browser, no install needed.
- 2Enter your values. The fields come pre-filled with realistic defaults so you can see how it works — replace them with your own numbers.
- 3Read the result. The output updates instantly. Copy or share it — nothing is uploaded to a server, everything stays on your device.
Frequently asked questions
Common questions about the A/B Test Sample Size Calculator.
Why do I need to calculate sample size for an A/B test?
Too few visitors and the result is statistical noise; this tells you how many you need per variant to detect a real difference with confidence.
What is Minimum Detectable Effect (MDE)?
The smallest improvement you want the test to reliably detect. A smaller MDE catches subtler wins but requires a much larger sample.
What is baseline conversion rate?
Your current conversion rate before the test — the control’s performance that the variant is measured against.
What does the confidence level mean?
The probability that a detected difference is real and not down to chance. 95% is the common default, meaning a 5% false-positive risk.
What statistical test does this use?
A two-proportion z-test, the standard method for comparing conversion rates between two groups.
Community rating
Discussion (0)
No comments yet. Start the discussion.
Keep exploring
Related tools across 7BusyBoss — all free, all instant.