Recursive SPF with the 10-lookup limit, DKIM keys and DMARC policy
A/B Test Calculator
Enter recipients and conversions (opens or clicks) for each variant to see whether the difference is statistically significant, with a p-value and a confidence interval for the difference. Plan the required sample size before sending.
Workspace
Variant A (control)
Variant B (test)
Sample size planner
How the result is calculated
Two-proportion z-test (two-sided): p̂ = (xA+xB)/(nA+nB), SE = √(p̂(1−p̂)(1/nA+1/nB)), z = (pB−pA)/SE, and p-value = 2·(1−Φ(|z|)). The difference is significant if p < 1 − confidence level.
Confidence interval for the difference (unpooled Wald): (pB−pA) ± z·√(pA(1−pA)/nA + pB(1−pB)/nB). Sample size per variant = (zα/2√(2p̄(1−p̄)) + zβ√(p1(1−p1)+p2(1−p2)))² ÷ (p2−p1)², where p2 = p1(1+lift) and p̄ is their mean; α = 5% two-sided.
FAQ
What does p = 0.03 mean?
If there were no real difference, the chance of seeing a gap this large or larger by luck is 3%. It is not the probability that B is better.
Should I measure opens or clicks?
Clicks or conversions. Opens are inflated by Apple Mail privacy protection, which makes them a weak judge even for subject-line tests.


