Promptmkr

GraderExamples › Conversion

"Give me some headlines" — the prompt with no way to pick a winner

A headline prompt that returns twenty options and no way to choose, plus the version that ranks them against a real criterion.

The prompt, as typed

Give me some headlines for my landing page. Make them catchy.

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

10/100 · grade F

Clarity of context0
Structure3
Specificity2
Reusability0
Decision logic0

Rewritten

90/100 · grade A-

Clarity of context 10 +10
Structure 6 +3
Specificity 6 +4
Reusability 8 +8
Decision logic 5 +5

The rewritten prompt

You are a conversion copywriter.

Product: {{product}}
Audience: {{audience}}
What they are currently doing instead: {{status_quo}}
Main objection: {{main_objection}}

Write 8 landing page hero headlines, max 12 words each.

Then rank the top 3 by:
1. How immediately clear the offer is to someone who has never heard of us
2. How directly the headline neutralises {{main_objection}}

Constraints: no questions as headlines, no "unlock", "revolutionise",
"supercharge", no wordplay that requires a second read.

For the top pick, write one line on what it assumes about the reader.

Why the rewrite works

A list of twenty catchy options moves the work back to you. Naming two explicit ranking criteria means the model has to defend a choice, and the defence is usually more useful than the headlines.

Common questions

Why does this prompt score badly?
It scores 10 out of 100 (F) because it leaves the model to decide 4 things you already know: what the page is selling and to whom; how many headlines and how long; what makes one headline better than another here; the objection the headline has to survive.
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