Verified against ChatGPT · 2026-08-10
Write feature bullets that hold up against the specific skeptic reading your listing, not a generic shopper
Produces five feature bullets each built to survive one real objection a skeptical buyer would raise, rather than five feature statements that only work if the reader is already convinced.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Write five feature bullets for Denveo memory foam office chair aimed at the most skeptical realistic version of my buyer — someone who has been burned by an overhyped listing before and reads bullets looking for the catch, not the pitch.
FEATURES TO COVER
Adjustable lumbar support, breathable mesh back, 5-year frame warranty, weight-tested to 350lbs, tool-free assembly
WHAT SKEPTICS SPECIFICALLY DOUBT ABOUT THIS PRODUCT CATEGORY
Office chair buyers have been burned by 'ergonomic' chairs that looked adjustable in photos but had lumbar support that didn't actually move, and warranties that excluded the part that broke
PROOF POINTS AVAILABLE
Frame load-tested by an independent lab to 350lbs static load; warranty explicitly covers frame, gas lift, and armrests, not just fabric
For each of the five features, write the bullet in this shape: a bolded 2-5 word claim tag, then one sentence that answers the specific skeptical question a doubtful reader would silently ask about that exact claim — not a generic elaboration, the actual doubt. If I gave you a real proof point (a test result, a spec, a certification) that supports the claim, use it; if I didn't give you one for a particular feature, write the bullet honestly with a softer, defensible claim rather than inventing a number or certification to sound more convincing.
Before the five bullets, name in one line what you believe the single most common silent objection is for this category, based on what I told you about category skepticism — this is what each bullet is being written to survive.
WHAT NOT TO DO
Do not write a bullet that is just the feature restated with an intensifier ("incredibly durable" is not an answer to "will this actually last"). Do not use unverifiable superlatives ("the best," "unmatched") anywhere — a skeptical reader treats an unverifiable superlative as a red flag, not reassurance, and it will cost more trust than it buys attention.
OUTPUT FORMAT
One line naming the core skeptical objection, followed by the five bullets, followed by a short note flagging any bullet where you had to soften the claim because no proof point was given for it.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
GPT-5.1's default mode for a bullet list is enumerative and additive — restate the feature, attach a positive adjective, move to the next one — which produces bullets that sound fine to a neutral reader but do nothing for the specific reader who has already been disappointed by similar claims before; naming the category-level skepticism explicitly and requiring each bullet to answer the silent doubt behind it forces the model to write toward a specific reader's mental objection instead of toward the feature itself. The instruction to soften a claim rather than invent a proof point when none was given directly counters a well-documented tendency of language models under a persuasive-writing brief to manufacture specificity (a percentage, a lab name, a certification) because specific-sounding claims score as more convincing in the model's training signal, even when nothing in the prompt supports them — for ecommerce copy this isn't just a style problem, it's a claim a merchant could be held legally accountable for making. Banning unverifiable superlatives matters mechanically for the stated skeptical reader specifically: a shopper primed to distrust the listing reads "the best" as evidence the seller has nothing more specific to say, which makes the superlative actively counterproductive rather than neutral filler, unlike with a more credulous reader where it might pass unnoticed.
What you get back
Core objection: that 'adjustable ergonomic' features are cosmetic and don't actually move once you sit down. **Lumbar support that actually slides** — the pad moves up to 3 inches on a separate track from the recline mechanism, so it stays where you set it instead of resetting when you lean back...
Verified against
ChatGPT GPT-5.1 · 2026-08-10
Changelog
- 2026-08-10 — 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
