SOCIAL

Twenty hooks, ranked by why they work

Use when the post is written but the first line is flat and you want options judged rather than just listed.

hook-generator-and-rank.md

Download .md

You are a copy editor who specializes in opening lines for social posts.

Post body:
"""
{{POST_BODY}}
"""
Platform: {{PLATFORM}}
Audience: {{AUDIENCE}}

Write 20 alternative opening lines. Each must be true to the body - no hook that promises something the post does not deliver.

Use a spread of these types and label each: specific number, named mistake, contrarian claim, before/after, overheard line, cost of inaction, direct address to a role, plain statement of result.

Then output a table with columns: Rank | Hook | Type | Why it works | Risk.
Rank by this rule, in order of priority:
1. Does it survive with no image and no context.
2. Does it name something specific rather than a category.
3. Would a knowledgeable reader disagree or nod hard.
Ties broken by shortness.

Banned: rhetorical questions, "Most people get this wrong", "Nobody talks about", "Unpopular opinion", "Let that sink in", emoji, em dashes, and any hook that could open a post on a completely different subject.

If fewer than 20 hooks can be written honestly from this body, write fewer and say why.

Fill in before running

Replace each placeholder with your own detail. The more specific you are, the less the model invents.

{{POST_BODY}}
{{PLATFORM}}
{{AUDIENCE}}

Getting a better result

The Risk column is the useful one - it catches hooks that overpromise before your audience does.
Paste your last three best-performing posts above the body and ask it to match their register.
Reject any hook that would work for a competitor's post too; that is the test for specific.

Questions about this prompt

When should I use this instead of just asking for ten opening lines?

When the body holds up and only the first line is flat. Asking for ten options and picking a favourite selects for what sounds good rather than what survives a feed. This one ranks against stated criteria, standing alone first, and attaches a risk to each hook so you can see what it would cost you.

What do I need before running it?

The finished {{POST_BODY}}, the platform and the audience. The tips suggest pasting your last three best-performing posts above the body so the register matches yours rather than a generic social voice. Without a real body the ranking cannot check whether a hook promises something the post never delivers.

What comes back, and which column matters?

Twenty labelled hooks, then a table ranked on surviving without an image, naming something specific, and provoking agreement or argument. The Risk column is the one to read. It catches the hooks that overpromise, and rejecting them there is much cheaper than learning it from the comments under a post that underdelivers.

What is the mistake here?

Choosing the boldest hook rather than the one your body pays off. Apply the test in the tips: if a hook would work just as well over a competitor's post, it is not specific enough to be worth using. Let it return fewer than twenty as well, since that is the signal your body carries one idea.