A dozen headline variants engineered for testing, not guessing.
prompt.txt
Act as a conversion copywriter. The asset: [landing page / ad / email subject / video title]. Current headline: [paste it]. What I sell and to whom: [one line]. Constraints: [character limits, tone rules].
Run the lab:
1. Score my current headline on the 4 U's: useful, urgent, unique, ultra-specific. Show the weak spot.
2. Generate 12 variants, 2 each across these levers: outcome-with-number, how-to, question, contrarian, proof-led, curiosity. Under [limit] characters each.
3. For each variant: one line on the hypothesis it tests — so results teach me something, whichever wins.
4. Pick your top 3 and defend them against each other in 2 sentences each.
5. Design the test: traffic split, the minimum sample before I look, and the decision rule (what margin declares a winner). Warn me about the classic mistake of calling tests early.
#copywriting#testing
💡 Replace the [brackets] with your details — the more specific, the sharper the answer.
Act as a short-form content strategist whose hooks average 3x watch-through. My topic: [topic]. Platform: [LinkedIn / X / Instagram / YouTube]. My audience: [who they are and what they secretly want].
Generate 20 hooks:
- 5 curiosity gaps (withhold the key word on purpose).
- 5 contrarian takes on a belief my audience holds.
- 5 specific-number leads (the number must be plausible, not clickbait).
- 5 story-openers that start mid-action.
Rules: every hook under 12 words where possible; no 'You won't believe'; no questions a reader can answer with 'no' and scroll past.
Then rank your own 20: top 3 with one line on the psychology each exploits. Finally, write the second line for your #1 pick — the line that keeps them after the hook did its job.
A brutal, useful teardown of the page your sales depend on.
Act as a conversion rate optimizer who has run 500+ landing page tests. Here is my landing page copy: [paste the full page text — headline, subhead, bullets, CTA, everything]. What I sell and to whom: [one line].
Roast it, then fix it:
1. The 5-second test: based only on my headline, what would a first-time visitor think I sell? Be blunt about the gap with reality.
2. Clarity killers: every phrase that is clever to me but confusing to them, with a plainer rewrite.
3. The missing elements: proof, specificity, risk reversal, urgency — which are absent and where exactly each belongs.
4. Rewrite my headline 5 ways, each on a different psychological lever (outcome, pain, curiosity, proof, direct).
5. Rewrite my CTA so it states what they get, not what they do.
6. The single highest-impact change if I only make one — and why.
A month of posts planned in one sitting, matched to your goals.
Act as a content strategist. My business: [what you sell]. Platform(s): [where you post]. Posting capacity: [posts per week you can honestly sustain]. Goal for the month: [followers / leads / sales / authority]. Content pillars I think I have: [topics, or 'help me define them']. My proof and stories: [results, opinions, behind-the-scenes material you can draw from].
Build my 30-day calendar:
1. First, define or refine my 4 content pillars and the percentage mix (e.g. teach / prove / relate / sell).
2. Then the calendar as a table: day, pillar, format, hook idea, CTA. Batch similar formats together for production efficiency.
3. Mark which 5 posts are the 'spend extra effort' posts and why.
4. Include one weekly repurposing move: how each week's best post becomes 2 more pieces.
5. End with the 3 metrics that tell me at day 30 whether the calendar worked — with the numbers that mean continue, adjust, or pivot.