Calculators
A/B Test Significance Calculator (z-test)
A/B Test Significance — 100% free, unlimited, no sign-up cost.
Enter visitors and conversions for variants A and B and the calculator returns a z-score. A value above about 1.96 means the difference is significant at roughly 95% confidence.
Compare two conversion rates with a pooled two-proportion z-test. It tells you whether the gap is likely real or likely noise, using only the visitor and conversion counts you enter.
Use the A/B Test Significance online at no cost.Common tasks this tool helps with
- a/b test significance formula
- how to calculate a/b test significance
- free a/b test significance
- a/b test significance with example
Try A/B Test Significance free
The Statistical Significance (z-test) formula
z = (p1 − p2) ÷ √(p̄(1−p̄)(1/n1 + 1/n2))
p1 and p2 are each variant's conversion rate, n1/n2 are visitor counts, p̄ is the pooled rate. A z-score above 1.96 is roughly 95% confidence.
Example
Variant A converted 5.2% of 2,000 visitors, Variant B converted 6.1% of 2,000 visitors.
pooled two-proportion z-test = a z-score telling you whether the 0.9pt gap is statistically real or noise
Worked example: 5.0% vs 6.2%
Variant A: 5,000 visitors, 250 conversions (5.0%). Variant B: 5,000 visitors, 310 conversions (6.2%).
- The pooled rate is 5.6%, giving a standard error of about 0.46 percentage points.
- z = (6.2% − 5.0%) ÷ 0.46% ≈ 2.6, above 1.96.
- The difference is statistically significant at roughly 95% confidence.
What to do next: Confirm the test ran its planned length with stable traffic, then check the lift matters commercially. Significance says the difference is unlikely to be chance, not that it will repeat.
Why it's free
A higher conversion rate on one variant does not prove it is better; small samples swing a lot. A significance check prevents shipping a change because of random variation. It cannot fix a badly run test, such as stopping early or changing traffic mid-test.
What you get
- A z-score for the difference between the two variants
Who uses the A/B Test Significance
- CRO teams reviewing test results
- Marketers testing landing pages or emails
- Founders deciding whether to keep a change
How it works
- Enter visitors and conversions for variant A
- Enter visitors and conversions for variant B
- Read the z-score: above about 1.96 is significant at roughly 95%
Frequently asked questions
Is the A/B Test Significance free?
Yes. Every one of GrowthGadgetAi's 100+ Marketing Toolkit tools, including the A/B Test Significance, is completely free with unlimited use — no credits, no plan requirement and no credit card.
Do I need to create an account to use it?
No. It runs on this page, in your browser, without signing in. Nothing you enter is sent to a server.
How is the A/B Test Significance's result calculated?
It's a direct formula, not an AI guess — the A/B Test Significance computes your z-score (>1.96 ≈ significant) from the numbers you enter, instantly and the same way every time.
How do I use the A/B Test Significance?
1) Enter your Variant A visitors and Variant A conversions and Variant B visitors and Variant B conversions. 2) See your z-score (>1.96 ≈ significant) instantly, no submit button needed. 3) Use the number in your reporting or planning. The whole process takes about a minute and runs on this page, on desktop or mobile.
What is the formula behind the A/B Test Significance?
Statistical Significance (z-test) = z = (p1 − p2) ÷ √(p̄(1−p̄)(1/n1 + 1/n2)). p1 and p2 are each variant's conversion rate, n1/n2 are visitor counts, p̄ is the pooled rate. A z-score above 1.96 is roughly 95% confidence.
Does a z-score above 1.96 mean variant B will keep winning?
No. It means the observed difference is unlikely to be random at about 95% confidence. Seasonality, novelty effects and traffic changes can still shift results later.
More free Calculators tools
Browse all 100+ free tools
Go further with AI Marketing Analytics & Reporting Tool: Move from campaign numbers to the next decision.
Read more: landing page optimization
Related tools