How do you compute sample size for A/B?

medium

Answer

${\sigma}^{2}\; / \;(\mathrm{MDE})^{2}$
${\sigma}^{2}$
$(2) minimum detectable effect MDE.$
$(3) significance α.$
$(4) power (1 - β).$
$Typical: 5% relative lift on a 10% baseline metric with p=0.05, 80% power needs 10K-100K users.$
$Reduce sample via CUPED / stratification.$
$Longer runs help capture weekly patterns but don't reduce sample requirement below noise.$

Check yourself — multiple choice

Random
n ≈ 16 ${\sigma}^{2}\; / \;\mathrm{MDE}^{2}$ per arm for α=0.05, power=0.8; inputs: variance / MDE / α / β; typical 10K-100K for 5% lift; reduce via CUPED / stratification; longer runs = coverage not sample
Just guess
Not real

Power analysis: n ≈ 16σ²/ $\mathrm{MDE}^{2}$ ; reduce via CUPED.

#experimentation

215 interview questions in this topic.