Writing prompt

Newsletter Writing Prompt — Turn Rough Notes Into a Tight Draft

Newsletters live and die on personal voice. Most AI writing prompts produce generic newsletter-speak — ‘In today’s rapidly evolving landscape…’ — that tells your subscribers a bot wrote it. This template forces structure and bans the cliches, so you keep your voice while saving 30 minutes of drafting.

✓ Works in ChatGPT, Claude, Gemini ✓ Tested template ✓ Copy + paste ready

The template

Replace the [bracketed] fields with your specifics. Send to any AI chat.

Below are rough notes from my week. Turn them into a 400-word newsletter.

Voice: first-person, plain English, one image or analogy, end with a single question to the reader.

Structure:
- Hook (1 sentence — concrete moment, not a question)
- Main idea (3-4 sentences)
- 1 supporting story or example (3-4 sentences)
- Takeaway in one line
- Closing question

Banned phrases: "In today's world", "It's no secret that", "Let me ask you this", "Imagine if you could", "Here's the thing"

NOTES:
[paste your raw notes here]

How to use it

Brain-dump your week’s notes — a few specific moments, an observation, a tool you tried. The messier the better; the prompt structures them.
Adjust the word count if your audience expects longer or shorter (300-600 is the typical newsletter sweet spot).
Always edit the output before sending. The prompt gets you 80% of the way; the last 20% needs your taste.

Example output

Here’s an example of what a quality AI returns when you run this template:

**Subject: The thing I almost canceled this week** Last Tuesday I almost canceled my Claude subscription. I’d been using it for three months for our newsletter drafts and had hit a wall — every piece felt too tidy, too packaged. The kind of writing that sounds correct but you can’t remember reading. Then I tried something different. Instead of asking it to draft from scratch, I gave it my rough Monday morning notes and asked it to *structure* them — keep the voice, just fix the order. The piece that came out the other side sounded like me, but better organized than I’d usually write. The pattern: AI is worst as a writer and best as an editor. Use it to cut, sort, and tighten — not to generate. What’s the AI workflow you almost gave up on but actually fixed?

Why this prompt works

Hook constraint as ‘concrete moment, not a question’.

Questions to the reader are the default opener for AI-generated newsletters and the biggest tell. Forcing a concrete moment makes the opening feel human.

Banned-phrase list is doing more work than the structure.

Three months of running this prompt and the banned-phrase list is what most reliably eliminates the ‘AI smell.’ Add your own pet peeves.

One image or analogy, not three.

Models love piling on analogies. Capping it at one forces them to pick the strongest.

Closing question is the conversion driver.

Newsletters that ask questions get replies. Replies are the highest-engagement signal email providers track.

Which AI to use

**Claude** wins this one decisively per writer-focused reviews. Its prose has more sentence-length variation and fewer LinkedIn-isms by default. **ChatGPT** can match it with strong custom instructions. **Gemini** tends to flatten the voice — usable but rarely first-choice for writing.

Read the full comparison in ChatGPT vs Claude vs Gemini in 2026 →

Frequently asked questions

Which AI writes the best newsletters?

Per writer-focused publications (Substack reviews, Every, writer-specific Wirecutter and Tom’s Guide pieces), Claude is the consensus pick for newsletter and long-form writing. ChatGPT is fine with custom instructions. Gemini trails on prose voice.

How long should an AI-drafted newsletter be?

300-600 words for most general-audience newsletters; 800-1500 for in-depth analytical newsletters. Cap the AI at your target length explicitly or it will overshoot.

Can I publish AI-drafted newsletters as-is?

Don’t. Even a great draft needs your edit pass to fix specific factual claims, sharpen your hooks, and re-insert opinions. Plan on 5-10 minutes of edits per AI-drafted piece.

Will Google penalize AI-drafted newsletters?

Newsletters live in inboxes, not search results — Google’s ranking algorithm doesn’t apply. The question is whether your subscribers detect AI; the banned-phrase list helps avoid this.

Can the AI write the subject line too?

Yes — add ‘Also give me 5 subject line options, under 7 words each, no clickbait’ to the prompt.

Build any prompt in 30 seconds

Free tool, no signup, runs in your browser. Pick a category, fill in the blanks, copy your prompt.

Try the prompt generator → Read the full guide