№03 · Use Case · Merchandising / Marketing
Hero Product Finder
Find the 1–2 pieces that can carry your brand.
Paid once. Checkout by Polar, our merchant of record, which works out any sales tax or VAT.
In these sets: Launch a Drop, $19 · All-Access, $49
Or included in membership.
- Format
- Use Case
- Made for
- Solo founders, small brands
- Difficulty
- Intermediate
- Time to value
- ~45–60 min
Times are estimates until a platform test run confirms them.
CompatibleChatGPT · Claude · Gemini · Grok · Any model
Overview
- The problem
- “We have 30 SKUs and spread our budget evenly. We don’t know which piece customers actually come back for, or what to build around it.”
- What you get
- A ranked table and a shortlist of 1–2 hero candidates with the evidence for each (sell-through, repeat buys, review themes), a confidence rating and watch-outs, an extension plan (colors, sizes, adjacent pieces) and a story angle, plus what data would firm up the call.
- What’s included
- Use-case brief (README + guide)
- Hero scoring rubric (sell-through, repeat rate, margin, review sentiment, return rate)
- Data template (CSV + column guide)
- Spreadsheet scorer (.xlsx, formulas, confidence, watch-outs)
- Extension plan template
- Story-angle template
- 3 starter prompts (+ full prompt set)
- Review checklist
- Platform setup notes (ChatGPT, Claude, Gemini, Grok Bot, any chat tool)
- Ready-made skill (SKILL.md folder + zip)
- PDF guide
- Illustrative example
- What you’ll need
- 6–12 months of sales by style-color (units received, units sold, revenue, returns)
- Buyers and repeat buyers if available
- Margin % if known
- Review counts (+ optional anonymized review text)
- Rough stockout and promo days
- Range list
- Brand bible (optional)
How it runs
7 steps- Fill in the template
- Bot checks data size and states its confidence
- You confirm the dataYour call
- Bot scores each SKU, shows the math and shortlists 1–2 candidates with evidence
- You sanity-check against what you know (stockouts, promos, seasonality)Your call
- Bot drafts extension plan + story angle
- You decide what to invest inYour call
Your call marks a step where you review and decide. Nothing moves on without you.
Estimated time: ~45–60 min including data prep.
Starter prompts
- Score my SKUs for hero potential
- Why is {SKU} a hero candidate, and what could be skewing it?
- Draft an extension plan for {SKU}
Fill in the parts in braces. The full prompt set is in the kit.
Sample output
Candidate 1: Merino Cardigan, Oat (hero score 4.42). Top in sell-through (87%) and repeat buyers (23%). 28 of 31 reviews positive. Themes: softness, ‘wears it every day’. Confidence: MEDIUM (8 months, a 21-day stockout, margin [MISSING]). Extension: sage first, then charcoal. Test a cropped length. Oat beanie in the same yarn. Story angle: ‘the one sweater you’ll actually wear every day.’
Produced by a manual walk-through of the prompt, not a platform test. It will be replaced with real output after a test run.
Platforms
| AI tool | Label | Setup note |
|---|---|---|
| ChatGPT | Compatible | Project + data analysis |
| Claude | Compatible | Project or skill, code execution for CSVs |
| Gemini | Compatible | Skill, attach the data in chat |
| GrokRuns as a Grok Bot (beta) | Compatible | No limits noted |
| Any modelAny capable chat model | Compatible | No limits noted |
We ran it ourselves on this tool. Shows the date and model of the run.
No piece carries this yet. Tested means we have run it ourselves. We update the label as each piece is tested.
Built to run on this tool. Not yet confirmed by our own test run.
Works with a limit we name.
Tools required
Any chat model with file upload or code/data analysis (or paste-as-text). A spreadsheet app for the scorer. Store/POS export. ChatGPT, Claude, Gemini and any-chat-tool setups work on free tiers. Grok Bot needs a SuperGrok or Cursor plan (beta).
House standards for this piece
All our standards- Human review gate
- You confirm the data, sanity-check the shortlist and make every investment decision.
- Data grounding
- Your own sales, returns, margin and reviews only. The math is shown and the confidence is stated.
- No invented results
- Never says a piece will become a hero and never predicts sales.
- AI disclosure
- Story angles are AI-assisted drafts. Edit them before any customer sees them.
- IP & bias
- Doesn’t copy competitor icons. Small data and stockouts can skew rankings, and it says so.
- Privacy
- Remove customer names, emails and order numbers from review exports.
- Approval boundary
- Recommends only. Never changes stock, prices, listings or ads.
Questions people ask
How much data do I need?
Six months is a workable minimum, and 12 months is better, so you can see a full year of seasons. With less, the bot will still score your pieces but will say its confidence is LOW.
Which AI tool should I use?
Whichever you already use. ChatGPT (Project), Claude (Project or skill), Gemini (skill), Grok Bot and any chat tool are all marked compatible: they should work, but we haven’t completed verified test runs yet. ChatGPT, Claude, Gemini and the any-chat-tool route work on free plans. Grok Bot needs a SuperGrok or Cursor plan.
Has this been tested on real brands?
Not yet. It needs testing on real fashion data. The rubric is a careful starting point, not a proven model, which is why every result comes with a confidence level and you sanity-check it against what you know.
I don’t know my margins. Can I still use it?
Yes. The bot leaves margin out and re-balances the other weights. Adding margin later (the Margin Checker kit can work it out) makes the call firmer.
Can I trust the bot’s math?
Check it. AI tools can slip on numbers, which is why hero-scorer.xlsx is included. If they disagree, trust the spreadsheet.
Will my hero definitely sell more if I extend it?
We can’t promise that, and neither can the bot. It shows you where the evidence points. You decide whether to invest, test small or wait.