Read the test against the decision you planned
Check integrity, effect size, uncertainty, guardrails, and time patterns before choosing to ship, stop, iterate, or remain inconclusive.
Skip this if Observational before-and-after comparisons without controlled assignment
Read a result against the decision rule written before the test. Do not start with the green badge, the p-value, or the segment that looks most exciting.
The American Statistical Association states that a p-value does not measure effect size, commercial importance, or the probability that the tested hypothesis is true. Business decisions should not rest only on whether a threshold was crossed.[1]
The rest of this lesson is available once the Atlas is unlocked.