All Tools
Core Tools
Margin & Markup
ROI
Churn & LTV
Break-Even
Invoice
SaaS & Startup
CAC & Payback
MRR/ARR Projection
SaaS Valuation
Finance & HR
Hourly Rate
Employee Cost
LLC Fees & Taxes
Marketing & Sales
ROAS
Commission
A/B Test
A/B Test Statistical Significance
Don't ship on a hunch. This runs a two-proportion z-test on your control and variant to tell you if the difference is real or just noise.
Control — Version A
Variant — Version B
Two-proportion z-test, 95% confidence threshold.
Statistically significant — or not?
A lift in conversion rate can be luck. The z-test measures whether the difference between two proportions could plausibly be zero. With a p-value below 0.05 (95% confidence), the result is usually treated as statistically significant.
How it's calculated
- Pooled rate = (A + B conversions) ÷ (A + B visitors)
- Z-score = (rate B − rate A) ÷ standard error
- p-value = two-tailed probability of that Z
- Confidence = (1 − p-value) × 100