A/B Testing
A/B Test Lift Calculator
Relative and absolute lift between control and variant conversion rates, plus projected extra conversions.
Relative lift = (B − A) / A × 100
Also called conversion lift · relative improvement
Marketing & Growth · 7 calculators
A/B Testing Calculators
A/B test lift, statistical significance, sample size, confidence intervals, test duration and minimum detectable effect.
A/B Testing
Sample Size Calculator
How many survey responses or A/B test users you need for a given confidence level, margin of error and population.
n₀ = z²·p(1 − p)/e²
Also called survey sample size · ab test sample size
A/B Testing
Statistical Significance Calculator
Is your A/B test result significant? Two-proportion z-test with p-value, confidence and relative uplift.
z = (p̂B − p̂A) / √(p̂(1 − p̂)(1/nA + 1/n…
Also called ab test significance · p value calculator
A/B Testing
Confidence Interval Calculator
Confidence interval for a mean (from mean, SD and n) or a proportion (from successes and n) at 90, 95 or 99% confidence.
Mean: x̄ ± z·σ/√n
Also called 95% confidence interval · margin of error calculator
A/B Testing
Conversion Difference Calculator
Compare two conversion rates from raw visitors and conversions: absolute and relative difference.
Δ = rate B − rate A
Also called conversion rate difference · compare conversion rates
A/B Testing
A/B Test Duration Calculator
How many days to run an A/B test to reach the required sample size, given daily traffic and number of variants.
n = (z_α/2·√(2p̄q̄) + z_β·√(p₁q₁ + p₂q₂))…
Also called ab test length · how long to run ab test
A/B Testing
Minimum Detectable Effect Calculator
The smallest relative lift an A/B test can reliably detect for your baseline rate, traffic, significance and power.
MDE ≈ (z₁₋α/₂ + z₁₋β)·√(2p(1 − p)/n)
Also called mde calculator · detectable lift
Key a/b testing formulas
| A/B Test Lift | Relative lift = (B − A) / A × 100 |
|---|---|
| Sample Size | n₀ = z²·p(1 − p)/e² · n = n₀ / (1 + (n₀ − 1)/N) |
| Statistical Significance | z = (p̂B − p̂A) / √(p̂(1 − p̂)(1/nA + 1/nB)) |
| Confidence Interval | Mean: x̄ ± z·σ/√n · Proportion (Wilson): (p̂ + z²/2n ± z√(p̂(1−p̂)/n + z²/4n²)) / (1 + z²/n) |
| Conversion Difference | Δ = rate B − rate A |
| A/B Test Duration | n = (z_α/2·√(2p̄q̄) + z_β·√(p₁q₁ + p₂q₂))² / (p₂ − p₁)² |
| Minimum Detectable Effect | MDE ≈ (z₁₋α/₂ + z₁₋β)·√(2p(1 − p)/n) |