PulseFit, a fitness subscription app, wants to test a new onboarding flow designed to increase paid signup conversion from landing-page visitors. Before launching the experiment, the growth team needs to know how many users are required for the test to be conclusive.
Estimate the required sample size per variant for a two-arm A/B test on conversion rate. The team wants enough traffic to detect a meaningful lift with high probability while controlling false positives.
| Metric | Value |
|---|---|
| Baseline conversion rate | 8.4% |
| Minimum detectable effect (relative lift) | 12.0% |
| Significance level | 5.0% |
| Statistical power | 80.0% |
| Test type | Two-sided |
| Traffic split | 50 / 50 |
| Weekly eligible visitors | 52,000 |
A 12.0% relative lift on an 8.4% baseline implies a treatment conversion rate of 9.408%.