CPG prompt of the day

Shopper study design: hypotheses before methodology

Someone wants to commission shopper research. Before money moves, force the hypotheses and check whether existing data already answers them.

You are an insights lead scoping a shopper study. I will describe the business question and what our existing data already shows. Produce:

A) FIVE testable HYPOTHESES, ranked by decision impact — each phrased so a result would clearly change a real decision (assortment, pricing, merchandising, media), not just be 'interesting'.
B) For each hypothesis, a short QUESTION TREE: what we would ask or measure, and what answer would change the decision.
C) A METHOD-FIT table ordered cheapest-first: which hypotheses existing panel or POS reanalysis could answer, which need a survey, which need in-store work (shop-alongs, intercepts) — with a one-line rationale each, and an explicit 'do not commission' call if existing data suffices.

Context: [DESCRIBE: the business question, the decision it feeds, what syndicated/panel/loyalty data already shows]

Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in the data I give you, write "not provided" and flag it. Mark every claim I should verify against my syndicated data or internal reporting before using it externally. Never include retailer-confidential terms or personally identifiable shopper data.

Why it works — Research budgets die on studies that were methodologically clean but decision-irrelevant. Ranking hypotheses by decision impact — and giving the model permission to say 'your existing data already answers this' — is the cheapest insight-quality upgrade a team can make.