A/B Test Calculator
Checks A/B test significance from visitors and conversions, shows the lift with a confidence interval, and calculates the sample size and duration you need.
| Variant | Conversion | Difference, p.p. | Growth | p-value | Verdict |
|---|
Enter visitors and conversions for each variant, and the calculator shows whether the difference is significant and how much the metric grew. The second tab estimates how many visitors a test needs and how many days it will take.
What the calculator measures
Significance is tested with a two-sided z-test for two proportions. The difference is shown two ways: in percentage points and as a percentage of the control group's conversion rate, each with its confidence interval. The interval matters more than the single number: a 30% lift with an interval from 2% to 66% means the real effect may turn out to be far more modest.
How to read the result
The p-value is the probability of seeing a difference this large or larger if the variants are in fact identical. It is not the probability that variant B is better. Run the test until you reach the calculated sample size: if you stop on the day the difference first becomes significant, false wins will be far more common.
Rate this tool
Help us get better
Tell us what was missing.
Other tools
Link Analyzer
Internal link analysis and a check of outgoing links on your site's pages.
Anchor Text Generator
A bracketed template expands into a set of phrases without repeats.
Sitemap Generator
A site map by crawling pages or from a ready list of addresses.
HTML Link Generator
Ready-made link code: a regular link, an email, phone or anchor.
Open Graph Checker
How your link looks when shared to social networks and messengers
Meta Tag Checker
Title, description and robots: what's in the code and what survives in results
Robots.txt Validator
Robots.txt errors and a check of whether a URL is blocked
Robots.txt Generator
A robots.txt for your CMS, with checks and a comment on every line