Product Story Framework
Two t-shirts. Same cotton. Same cut. One costs £15, the other £45 — and the £45 one sells better because it has a story. Story is the multiplier that turns product attributes into perceived value. The Product Story Framework structures the real story you already have into a five-layer narrative that works as a one-liner, an About page, a product page, and an ad.
What this skill does
True stories only. Fabricated origin tales get exposed and destroy trust faster than no story at all. The skill works with what's actually true — and if the real story is "I noticed a market gap and decided to build something", that's a perfectly valid origin. Don't manufacture a crisis you didn't have. Authenticity beats drama every time, and a quiet origin told well outperforms an invented epic.
The five-layer architecture is modular by design. A brand story that only works as a 500-word About page is useless — it needs to compress into an Instagram bio, expand into an investor pitch, and distil into a product-page paragraph. Layer 1 is the hook: one sentence that captures the essence and goes everywhere. Layer 2 is the origin in 3-5 paragraphs structured as before / moment / bet / proof. Layer 3 is the process — the corner that isn't cut, the care that's made visible. Layer 4 is the proof, structured as customer mini-stories rather than disembodied quotes. Layer 5 is the invitation — the CTA reframed as joining something rather than transacting. Each layer is built so it can be lifted out and dropped into the surface that needs it.
The narrative core is chosen deliberately, not picked from a menu. The Frustration Story ("I couldn't find X that did Y, so I built one") fits founder-led problem-solving brands. The Craft Story fits premium and artisan goods where the corner-not-cut is the differentiator. The Mission Story fits purpose-driven brands. The Discovery Story fits novel ingredients or methods. The Community Story fits brands that customers helped build. The skill picks based on the actual story rather than defaulting to the dramatic one.
Resellers and white-label brands get a curation story rather than a manufacturing one. If you don't make the product, the story is about finding and selecting — "we test 50 products to stock 1" is a legitimate narrative, and forcing a maker tale on a brand that doesn't make things ends in greenwashing's cousin: craft-washing. The skill recognises the difference.
Sustainability claims must pass the "prove it" test. "Eco-friendly" means nothing. "Every unit diverts 2.3kg of waste from landfill" means everything. If you can't put a number, a certification, or a verifiable claim behind the green statement, the skill recommends leaving it out — vague sustainability messaging is worse than none, and the backlash for greenwashing is real. The same rule applies to "ethically made" and "small-batch" — specific or removed. Output includes the full five layers, a deployment grid showing which layers go on which surface (homepage hero, About, product page, ads, email welcome, packaging, investor deck), and brand voice notes covering tone, words to use, and words to avoid.
When this triggers
- ·You have a strong product and a real backstory but no structured way to tell it across surfaces
- ·Your About page is three paragraphs of generic mission statement and converts no one
- ·You're a premium brand and need to justify the price with something more than feature copy
- ·You're a reseller or curator, not a maker, and need a story that fits without faking a workshop
- ·Your sustainability claims are vague ('eco-friendly') and you suspect that's hurting more than helping
Example
Trigger
User: 'Heritage knitwear brand. Founder learned the craft from her grandmother in Donegal. Made in a small mill in Ireland. £180 jumpers. About page reads like a press release.'
Output
Narrative core: Craft Story. LAYER 1 — HOOK "Knitwear from the mill that taught my grandmother — made the slow way, the way it lasts." LAYER 2 — ORIGIN (3 paragraphs) Before: the closure of small Irish mills through the 90s. Moment: finding [Mill Name] still working — three machines, six staff, fifty years of patterns in a filing cabinet. Bet: ordering a first run of 40 jumpers with the founder's savings. Proof: sold out in 6 weeks, waitlist of 200. LAYER 3 — PROCESS The corner not cut: each jumper takes 14 hours on the machine vs. 90 minutes in a fast-fashion factory. Yarn sourced from [region], spun in [town]. LAYER 4 — PROOF Two customer stories (transformation, not "5 stars"). Numbers with context. LAYER 5 — INVITATION + deployment guide: which layer goes where.
Get this skill + 13 more
Included in the The Startup Founder Stack — move fast with ai leverage. Save $150+ vs buying individually.
Get The Startup Founder Stack — $149What you get
- 174-line SKILL.md, ready to drop into ~/.claude/skills/
- Tested through 3 Karpathy-loop iterations (versions v1.0.0 → v1.3.0)
- Triggers automatically when relevant — no command to remember
- Lifetime updates as the skill is refined further
More from E-Commerce
Takes a product and cart context and produce a 3-email abandoned cart sequence — each with a different psychological trigger, timed for maximum recovery rate
Takes a product, audience, and platform and produce 10 ad variants — each using a different psychological angle — ready to test
Takes product details and produce a complete, algorithm-optimised Amazon listing: title, bullet points, description, A+ content brief, and backend keywords
Takes a business's customer data, product type, and purchase patterns and build a complete retention system: segmentation logic, win-back flows, loyalty programme design, and VIP tier architecture
Takes negative reviews and produce professional, empathetic responses that turn public complaints into trust-building moments
Designs the complete system for collecting, organising, and activating customer data that the brand owns directly — email addresses, purchase history, preferences, and behavioural signals
Browse the full library
297 skills across 31 categories. One purchase, lifetime updates.
See all bundles