A/B Test Calculator

Statistical significance calculator for conversion rate experiments

Uses: 0

Visitors

Conversions

Visitors

Conversions

Confidence Level

90% 95% 99%

Results

Variant A

Variant B

Relative Improvement

Statistically Significant?

P-Value

Confidence

Z-Score

Variant A

Variant B

Estimate how many visitors you need before running your test.

Baseline Conversion Rate (%)

Minimum Detectable Effect (%)

Confidence Level (%)

Daily Visitors (per variant)

visitors per variant required

Run tests for at least 1-2 full business cycles (typically 2-4 weeks) to account for day-of-week variation
Use a 95% confidence level as the industry standard; 99% for high-stakes changes
Decide your sample size before starting the test, not after peeking at results
Test one variable at a time for clear attribution of what caused the difference
Avoid stopping tests early when results look promising (peeking problem)
Ensure random, even traffic splitting between variants
Document your hypothesis before running the test
Account for novelty effects by running tests long enough for user behavior to normalize

Lifetime Pass ($99.99) gets every tool we build. Or resell them all under your brand from $99.