cohort-analysislisted

# The Cohort Tracker Build a cohort retention or revenue table: group customers by the period they were acquired in, then track a single metric across each period after that, so the user can see whether performance is improving or decaying cohort over cohort, not just watch one blended number drift. > **Chart form.** Read `references/chart-form-and-accessibility.md` before specifying how any > number is displayed. Its cohort table follows the colour-scale and legend rules there, and the requirement that a diverging scale is used only where a real midpoint exists rather than an arbitrary one. > **Input integrity.** Run the checks in `references/data-input-integrity.md` before computing > anything, and report what they found. Each one produces a confident wrong answer rather than > a visible error, so a broken input does not announce itself. Recent cohorts are immature rather than worse, and the most recent period is usually partial on both axes of the table. > Where a check cannot run because the export lacks the field, say so and state what it limits > the conclusion to. ## Before you write **Run the input list below before you write anything. If one of those inputs is missing, ask for it and stop. Do not return a draft with a warning on it.** The user copies the draft and leaves the warning behind, so a caveat protects you and not them. **Ask at most THREE questions. Hard cap.** Before anything becomes a question, get it yourself: read `.agents/product-context.md`, fetc