Promptmkr

GraderExamples › Ecommerce

"Write a product description" — the four gaps that make it generic

A product description prompt that produces adjectives instead of information, and the version that produces something a buyer can act on.

The prompt, as typed

Write a product description for my new product. Make it sound good.

What it leaves the model to guess

Graded before and after

Both versions below were scored by the same rubric the free prompt grader runs.

As typed

13/100 · grade F

Clarity of context5
Structure0
Specificity0
Reusability0
Decision logic0

Rewritten

90/100 · grade A-

Clarity of context 8 +3
Structure 6 +6
Specificity 8 +8
Reusability 6 +6
Decision logic 8 +8

The rewritten prompt

Write a product description for a {{channel}} listing.

Product: {{product_name}}
Specifications: {{specs}}
Buyer: {{buyer}}
Usually compared against: {{alternatives}}
Main hesitation before purchase: {{main_objection}}

Output:
- One-line summary, max 20 words
- 90-130 word description
- Four bullet points, each stating a specification and what it means for the buyer

Tone: {{tone}}. Channel: {{channel}} product listing.

Constraints: every claim must map to a specification listed above. No
"premium", "luxurious", "game-changing", or unverifiable superlatives.

Address {{main_objection}} once, directly, without naming it as an objection.

Write 3 versions of the one-line summary and rank them by how quickly a buyer
skimming a listing grid would understand what the product is. Say why the top
one wins. The description succeeds if every sentence would survive a buyer
asking "how do you know that?".

Why the rewrite works

The constraint that every claim must map to a listed specification is what stops the model inventing benefits. Without it, "make it sound good" is an invitation to write adjectives, and adjectives are what buyers skip.

Common questions

Why does this prompt score badly?
It scores 13 out of 100 (F) because it leaves the model to decide 4 things you already know: the product's actual specifications; who is buying it and what they compare it against; where the description will appear and how long it can be; the objection that stops people buying.
Does a longer prompt always score higher?
No. The rubric rewards named inputs, a fixed output format and an explicit rule for choosing between options. Extra prose without those adds length and no score.
Will this rewritten prompt work in ChatGPT and Claude?
Yes. It is plain text with placeholders you fill in before running, so nothing in it is specific to one model or vendor.

More worked examples

More bad prompt examples · How to write better AI prompts