the-returns-minerlisted

> **Validate reason codes against the product before trusting the concentration.** A code that cannot > apply to a SKU — "wrong size" on a non-apparel item, "damaged in transit" on a digital good — is > evidence of miscoding, and miscoding invalidates the root-cause conclusion even when the totals look > clean. Check that each SKU's top reasons are physically possible for that product, and where they are > not, report the coding problem as the finding rather than the reason as a cause. A catalogue where > staff pick the first dropdown option produces a confident wrong answer every time. > **When an input is missing, choose a response - never fill the hole silently.** Read > `references/missing-input-protocol.md`. Every absent input resolves to exactly one of **block** > (unsafe or non-compliant without it), **withhold** (print `withheld — <field> missing` where the > number would go), **degrade** (deliver a weaker honest version and name the tier), or **assume** > (state it inline at the point of use). There is no fifth option: never proceed as though the input > were present, never guess a number, and never drop the field so the gap becomes invisible. > > A required output field with no corresponding input is a defect in this skill, not in the user's data: > print it as `not supplied`, say what it would change, and ask for it once, specifically. # The Returns Miner Take a returns export with SKU-level reason codes and find which products have a real return concentration