A/B Test Sample Size

    1215

    Calculate required sample size for A/B tests with statistical power analysis

    NOTEModel Overview
    🧪 A/B TEST SIZING Determines required sample size: • Baseline conversion rate • Minimum detectable effect • Desired statistical power Uses standard power formula.
    VARIABLEBaseline Conversion
    beta
    Current conversion rate (~5%)
    baseline
    VARIABLERelative Lift
    uniform
    Minimum detectable effect (10-20%)
    lift
    CONSTANTStatistical Power
    0.8
    80% power (standard)
    power
    CONSTANTSignificance Level
    0.05
    5% significance (standard)
    alpha
    FORMULATreatment Rate
    baseline * (1 + lift)
    Expected treatment conversion
    baseline
    lift
    treatment
    FORMULAPooled Variance
    baseline * (1 - baseline) + treatment * (1 - treatment)
    Combined variance
    baseline
    treatment
    variance
    FORMULASample Size
    2 * variance * pow(1.96 / alpha + 0.84 * power, 2) / pow(treatment - baseline, 2)
    Users per variant (using power/alpha)
    variance
    alpha
    power
    treatment
    baseline
    sample_size
    OUTPUTRequired Sample
    Users per Variant

    About the creator

    The team behind Carlo. We believe everyone deserves tools to reason about risk and uncertainty.

    What is Carlo?

    Carlo is a visual tool for Monte Carlo simulation. Model uncertainty by dragging probability distributions, connecting them visually, and running thousands of scenarios instantly.