Experiment planning · No signup
A/B Test Sample Size Calculator
Estimate a fixed-horizon, two-variant sample size from your baseline rate and smallest worthwhile improvement.
Planning estimate
31,235
visitors per variant
63
estimated days
5.50%
target rate
Two-sided normal approximation, equal allocation. Use this for planning—not as a substitute for a pre-registered analysis plan.
How to use it
A useful estimate starts with honest inputs.
- 01Use a recent, representative baseline conversion rate
- 02Choose the smallest relative uplift worth acting on
- 03Keep confidence and power fixed before seeing results
- 04Account for weekly cycles and avoid stopping early
Turn the plan into measured results
TrailPixel connects privacy-first analytics, automatic funnels, and A/B testing.
Start free →