How much faster does CUPED make your A/B test?

Turn one number — the correlation between a pre-experiment covariate and your metric — into the variance reduction, effective sample-size gain, and runtime savings CUPED buys you.

Pearson correlation between the pre-period covariate (e.g. last week's value) and the in-experiment metric. Sign does not matter — CUPED uses ρ².
Optional. The variance of your metric without adjustment. Used to show the adjusted variance in the same units.
Runtime you'd need without CUPED. We scale it by the sample-size ratio to show the shortened schedule.
Variance reduction
30.3%
1 − ρ²
Adjusted variance
0.698
Var(Y)·(1−ρ²)
Effective sample ×
1.43×
1 / (1−ρ²)
Runtime saved
8.5 d
28 → 19.5 d

What CUPED actually does

CUPED (Controlled-experiment Using Pre-Experiment Data), introduced by Deng, Xu, Kohavi and Walker at Microsoft, removes the part of your metric that was already predictable before the experiment started. If each user has a covariate X from the pre-period, CUPED forms an adjusted metric:

Y_cuped = Y − θ·(X − E[X]),  where θ = Cov(Y, X) / Var(X)

Because θ is the least-squares slope, the adjusted metric keeps the same expected treatment effect — it is unbiased — but its variance drops to:

Var(Y_cuped) = Var(Y) · (1 − ρ²)

where ρ is the correlation between X and Y. That single ρ² term is the whole story. Variance in an A/B test drives how many samples you need, and required sample size scales linearly with variance. So the effective sample-size multiplier is 1 / (1 − ρ²), and — assuming traffic is roughly constant — the required test duration shrinks by the same factor. A covariate correlated at ρ = 0.7 cuts variance by 49%, nearly doubling your effective users and halving runtime; at ρ = 0.3 the payoff is a modest 9%.

The best covariate is usually the same metric measured on the same users in the period just before the test. Notice the reduction depends on ρ², so weak covariates barely help while strong ones pay off super-linearly. The chart above plots 1 − ρ² across the full correlation range and marks your current value, so you can see whether wiring up the pre-experiment join is worth the engineering before you build it. CUPED changes nothing about your randomization, false-positive rate, or the estimand — it is pure variance reduction, which is why it has become a default in mature experimentation platforms.