Verified against ChatGPT · 2026-08-08
Stress-test a new product idea against real demand signals before it becomes a launch plan
Produces a structured research brief evaluating whether a proposed new SKU idea is actually worth developing, weighing the demand signals and risks given rather than defaulting to an optimistic go-ahead.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Evaluate a proposed new product idea before it goes any further into development. The job here is an honest go/no-go/modify read based on the actual signals given, not a cheerleading brief that assumes the idea is good because someone proposed it. PROPOSED PRODUCT IDEA A travel-size version of the brand's bestselling 16oz body lotion, in a 2oz TSA-compliant bottle. WHERE THIS IDEA CAME FROM Recurring pattern in support tickets and Instagram comments over the last 4 months, roughly 30 separate mentions asking for a travel size. DEMAND SIGNALS AVAILABLE 30+ unprompted customer requests logged in support, plus the top competitor's travel-size version has 800+ reviews on its own listing. EXISTING CATALOG OVERLAP The existing 16oz bottle is the top seller in the skincare category; a travel size would likely be bought by the same repeat customers rather than new ones. CONSTRAINTS Supplier minimum order is 5,000 units for any new bottle mold, and there's no budget allocated for a new SKU until next quarter. STEP 1 — WEIGH THE DEMAND SIGNAL Given where the idea came from and the actual demand signals, state plainly how strong the evidence for real demand actually is — a single customer request is a much weaker signal than a recurring pattern across support tickets or search data, and the brief should say which kind of signal this is rather than treating any stated demand signal as automatically sufficient. STEP 2 — CHECK CATALOG OVERLAP Given the existing catalog, assess whether this would meaningfully expand what the store offers or largely cannibalize an existing product's sales — if overlap is high, say so plainly rather than treating a new SKU as pure incremental revenue by default. STEP 3 — CHECK AGAINST CONSTRAINTS Given the stated constraints (budget, minimum order quantity, timeline, or capability), flag anything about the idea as proposed that doesn't fit within them, and note what would have to change about the idea itself to fit — not just a restatement that the constraint exists. STEP 4 — VERDICT Give one of three verdicts — proceed, proceed with a specific modification, or don't proceed — with the single main reason. Do not hedge into "it depends" without naming what it actually depends on and what evidence would resolve that dependency. WHAT NOT TO DO Do not treat the fact that an idea was proposed as evidence it's a good idea — the brief's job is specifically to test that assumption, not confirm it. OUTPUT FORMAT Four labeled steps as specified, ending with the verdict stated in one clear sentence.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The explicit instruction not to treat a proposed idea as inherently good counters a strong default bias in how GPT-5.1 handles evaluation requests framed around someone's existing idea — asked to 'evaluate this product idea,' it tends toward a structurally polite, mostly affirming analysis that surfaces a token risk or two but ultimately validates the premise, because the prompt's framing implicitly signals that a positive answer is expected, and disagreeing with the premise of a request reads as an unhelpful response unless explicitly invited. Requiring the demand-signal step to distinguish types of evidence (a recurring documented pattern versus a single anecdote) rather than accept any stated demand signal at face value addresses a real failure mode in product decisions: 30 unprompted mentions across four months is meaningfully different evidence than one enthusiastic customer email, and a brief that doesn't force this distinction will happily build a confident-sounding case on thin evidence because the prompt handed it a 'demand signal' without qualifying its strength. The catalog-overlap check exists because new-SKU proposals are almost always pitched and evaluated as incremental revenue, when a travel-size version of an existing bestseller is a textbook cannibalization risk — the same repeat customers buying a smaller size instead of the full size, not new customers being reached — and a brief that skips this check will overstate the idea's actual revenue impact by counting cannibalized sales as new sales. Forcing a specific verdict rather than allowing an 'it depends' non-answer matters because product development discussions frequently stall in exactly this kind of unresolved ambiguity, and a brief that names precisely what evidence would resolve the dependency turns a vague hedge into an actionable next step.
What you get back
Verdict: proceed with modification — demand signal is genuinely strong (recurring documented pattern, not anecdote), but the 5,000-unit minimum order and cannibalization risk mean this should launch as a value-add with a full-size purchase (travel size as a paid add-on or gift-with-purchase) rather than a standalone SKU competing with the 16oz bestseller.
Verified against
ChatGPT GPT-5.1 · 2026-08-08
Changelog
- 2026-08-08 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
