A/B test duration calculator
Find out how many visitors each version needs and how many days your test must run before you can trust the result.
Example numbers. Replace them with yours.
- Days to run
- 29 days
- Visitors per variant
- 28,304
- Total visitors
- 56,608
- Conversions per variant
- 425
What it calculates
An A/B test needs enough visitors to tell a real difference from chance. The smaller the lift you want to detect and the lower your conversion rate, the more visitors you need. This calculator uses the standard two-proportion sample size formula at 80% power and your chosen confidence.
How to read the result
If the test needs months, test bigger changes. A larger lift needs far fewer visitors. Run at least one full week so every day of the week is included, and don't stop early because one version looks ahead.
The formula
p₂ = p₁ × (1 + lift)
n per variant = (z_α × √(2p̄(1 − p̄)) + z_β × √(p₁(1 − p₁) + p₂(1 − p₂)))²
÷ (p₂ − p₁)²
Days = n ÷ (daily visitors ÷ 2), at least 7Worked example
Example inputs
- Visitors per day
- 2,000
- Current conversion rate
- 1.5%
- Smallest lift to detect
- 20%
- Confidence
- 95%
Results
- Days to run
- 29 days
- Visitors per variant
- 28,304
- Total visitors
- 56,608
- Conversions per variant
- 425
What changes the result
Smallest lift to detect at five values. Every other input stays as in the worked example; the highlighted row is the example.
| Smallest lift to detect | Days to run | Visitors per variant |
|---|---|---|
| 10% | 109 days | 108,153 |
| 15% | 50 days | 49,193 |
| 20% | 29 days | 28,304 |
| 25% | 19 days | 18,519 |
| 30% | 14 days | 13,141 |
Use your real numbers
Analify reads daily sessions and conversion rate from Google Analytics, the two numbers this calculator needs.
Analify already reads 2 of these 4 inputs live, from Analytics
Start freeQuestions
How long should an A/B test run?
Until each version reaches the sample size, and at least one full week.
Why do I need so many visitors?
Small differences in a low conversion rate are hard to separate from chance. Detecting a bigger lift needs far fewer visitors.
What does 95% confidence mean?
If there were no real difference, a result this strong would show up by chance only 5% of the time.
Do you store what I enter?
No. The calculation runs in your browser and nothing is sent to Analify.
Give AI the business it was missing.
Connect your company, build its memory and start asking, deciding and approving from one place.
See pricing