Ecommerce & Shopify · 100% free

A/B Test Sample Size Calculator

How many visitors per variant before your CRO test reaches statistical significance?

Shopify WooCommerce Any CRO tool

Baseline conversion rate

Minimum detectable effectRelative lift e.g. 10% = 2.5% → 2.75%

ConfidenceStandard: 95

Statistical powerStandard: 80

Variants (incl. control)2 = A/B; 3 = A/B/C

Daily visitors in the test

Sample size per variant

64,200

Total across all variants: 128,399

Days to significance

65 days

At your daily visitor rate

Conversion rates being tested

2.50% → 2.75%

Detecting a 10% relative lift

Why this matters: 78% of A/B tests at small ecommerce stores are statistically meaningless — they're called too early. Calling a test before reaching this sample size is roughly equivalent to a coin flip on whether the winner actually wins in production.

Rule of thumb: if "days to significance" exceeds 4 weeks, your effect size is too small to detect at your traffic. Either test a bolder change, raise traffic, or skip the test.

About this tool

Most A/B tests on Shopify get called too early on too little traffic and end up making the store worse. This tool uses standard statistical-significance maths (95% confidence, 80% power) to tell you the per-variant sample size you actually need before declaring a winner.

Why MDE matters

Minimum Detectable Effect is the smallest lift you can reliably detect. A 1% MDE on a 2.5% baseline means you'd detect a winning variant that lifts conversion from 2.5% to 2.525% — that's an enormous sample size. In ecommerce, most tests should target a 10–25% MDE: detect meaningful moves, ignore noise.

When not to A/B test

If your math says you need 8+ weeks to reach significance, just don't test it. Either ship the change (when the downside is bounded), reserve testing for bigger-impact changes, or wait until you have more traffic. Underpowered tests waste time and lock teams into false conclusions.