E-Commerce & Product

Verified against ChatGPT · 2026-08-08

Turn a spec sheet into a product description that sells the outcome, not the feature list

Builds a product description by laddering every raw spec up to the concrete outcome it produces for one named buyer, so the page reads as a reason to buy instead of a restated data sheet.

ChatGPT (GPT-5.1)5 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

You are writing a product description page for an ecommerce listing. My biggest risk is writing a features list with adjectives sprinkled on top instead of copy that tells a specific buyer why this specific spec matters to them.

PRODUCT
Aria 3-in-1 stainless steel pour-over kettle

RAW SPECS OR FEATURES
0.9L capacity, gooseneck spout, thermometer built into lid, 304 stainless steel, induction-compatible base

TARGET BUYER
Home coffee hobbyists who already own a burr grinder but have been using a standard kettle and getting inconsistent pour speed

MAIN OBJECTION TO OVERCOME
They think a gooseneck kettle is a gimmick that won't actually change how their coffee tastes

BRAND VOICE
Direct and a little dry, no exclamation points, talks to the buyer like a knowledgeable friend not a salesperson

RULES
For every spec I gave you, ladder it up through "which means" until you reach a concrete, physical or emotional outcome the target buyer experiences — a spec that stops at "which means it's more durable" hasn't finished laddering; keep going until you hit something the buyer would actually notice in their life. Do not invent a benefit that isn't supportable by the spec I gave you — if a spec has no clear buyer-facing consequence, say so instead of manufacturing one. Open the description with the single outcome that most directly answers the main objection, not with a generic hook about the product category. Write in the specified brand voice but never let voice override clarity — a joke or flourish is cut before a fact is cut. Do not use the words "revolutionary," "game-changing," "premium quality," or "perfect for" — these are the words that make ecommerce copy indistinguishable from every competitor's.

WHAT NOT TO DO
Do not write a bullet list disguised as prose — this is a description, and it should read as 2-4 short paragraphs a real person would read before checkout, not a wall of comma-separated adjectives.

OUTPUT FORMAT
1. The finished product description (120-220 words).
2. A table mapping each raw spec to its laddered outcome, so I can see the reasoning and cut anything that feels like a stretch.
3. One line naming which sentence most directly addresses the main objection.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

GPT-5.1 defaults to a feature-forward register when it's handed a spec sheet, because specs are the most information-dense part of the input and the model naturally optimizes for including everything it was given rather than deciding what a buyer actually needs to hear — the explicit "which means" laddering instruction forces a second inference pass per spec instead of a single restatement pass, which is the mechanical difference between a spec sheet with better adjectives and copy that argues a case. Requiring the model to stop only when it reaches something the buyer would actually notice, rather than an intermediate abstraction like "more durable," closes the most common failure mode of AI-generated product copy: benefits that are technically true but still too abstract to move anyone, because the model stopped laddering one step too early. Banning specific words like "premium quality" and "game-changing" matters because these are exactly the phrases GPT-5.1 reaches for as a default intensifier when it doesn't have a concrete outcome to state instead — removing the escape hatch forces the substantive answer. The output table forcing a visible spec-to-outcome mapping exists because it's the only way to audit whether a claimed benefit is actually supportable by the given spec or whether the model quietly stretched — a failure mode transformer language models are especially prone to when asked to sound persuasive under a word-count constraint.

What you get back

Most kettles pour in one uncontrollable rush, which is why your first thirty seconds of blooming coffee grounds never come out even. The Aria's gooseneck spout narrows that rush into a thin, steady stream you can actually aim — so the water hits the grounds in the slow spiral your recipe actually calls for, instead of flooding one side of the filter...

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
All E-Commerce & Product prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY