How it works
Approximate sample per variant = 2 × p̄(1−p̄) × (zα + zβ)² ÷ absolute difference².
Input definitions
- Baseline conversion (%)
- Enter the value in the labeled unit.
- Minimum detectable absolute change (percentage points)
- Enter the value in the labeled unit.
- Two-sided confidence z value
- Enter the value in the labeled unit.
- Power z value
- Enter the value in the labeled unit.
A worked example
- Baseline conversion (%)
- 5
- Minimum detectable absolute change (percentage points)
- 1
- Two-sided confidence z value
- 1.96
- Power z value
- 0.84
Assumptions & limits
Uses only the values entered. Rates are illustrative inputs, not live quotes, legal rules or personalised advice. Check provider terms, local requirements and measurements before acting.
Sources & review
Maintained by AnsRio. Methods and examples use automated checks; this does not imply independent professional certification.
Method updated 2026-10-19. Content checked against the stated method and examples. Our verification approach.
Report an issue with this tool or suggest an improvement ↗
Common questions
Is this exact for every experiment?
No. Sequential testing, unequal allocation, multiple comparisons and very small rates need a more specific design.
How does Conversion-rate sample size calculate the result?
Approximate sample per variant = 2 × p̄(1−p̄) × (zα + zβ)² ÷ absolute difference².
What should I verify before using Conversion-rate sample size?
Uses only the values entered. Rates are illustrative inputs, not live quotes, legal rules or personalised advice. Check provider terms, local requirements and measurements before acting.