Turn best performing newsletter posts into unlimited new posts

Scan time: 3-4 min / Read time: 5-7 min

Hey rebel solopreneurs 🦸‍♀️🦸‍♂️

Most newsletter creators burn out writing fresh posts every single time.

Smart ones let AI audit their archive and multiply what works.

⛳️ Why this works

You've been publishing for years. Maybe hundreds of posts. Maybe thousands.

Every week? Same cycle. Blank page. Cursor blinking. "What should I write about?"

5 hours later. One new post. Published. Done.

Next week? Back to zero. Start over. Again.

Here's the thing:

Your newsletter archive is like an oil rig.

Most creators treat it like a scrapbook. Something to look back on. Maybe feel proud about. But never actually use.

Ryan Holiday has written multiple books about Stoicism. Thousands of articles. Hundreds of YouTube videos. All about Marcus Aurelius.

Same topic. Over and over. For years.

Why? Because volume on one topic is how you own a niche.

But wait. Writing the same thing repeatedly? Exhausting.

Even if you've written 1,000 posts, manually going through your archive to find what works, analyzing patterns, then rewriting variations?

That could take weeks.

Turns out, you don't need to do it manually.

If your posts are public, AI can search your entire archive. Find your top topics. Fetch your best posts. Learn your patterns. Generate variations in 60 seconds.

You spend 10 minutes cleaning it up. Publish.

That's content multiplication. Bingo.

Let's see how Maya figured this out:

📋 This works best for public newsletters. Got paywalled content? Just copy/paste 3-5 posts into Prompt 2 and skip Prompt 1.

Maya runs a Beehiiv newsletter. Been publishing since 2018.

Over 6 years? 1,000+ posts. All public. Some about productivity. Some about AI. Some about creator economy.

At her peak in 2020? $100K per year from her newsletter. Most of it from variations of her best-performing topics.

But here's her problem.

After 6 years and 1,000 posts, Maya was tired. Every new post meant starting from scratch. Blank doc. Cursor blinking.

She knew some topics performed better than others. But which ones?

She could manually review all 1,000 posts. Read each one. Track topics. Note patterns.

That would take weeks.

Or she could guess based on memory. "I think productivity posts do well?"

Unreliable.

Maya was tired of having a massive archive but no idea what actually worked.

Then she found something. A principle from newsletter creators who'd built million-subscriber lists.

A concept called "Public Content Multiplication."

It explained exactly why manual archive analysis was killing productivity. And how to let AI audit your entire library and generate variations in minutes instead of weeks.

Maya decided to follow these steps:

Step 1: Let AI search her newsletter and identify top topics

Step 2: AI fetches her best posts, learns patterns, and generates title ideas

Step 3: Pick a title and AI writes the full post

🔍 Step 1: Maya had AI audit her entire archive

Maya opened ChatGPT/Claude (her AI sidekick).

She had 1,000+ posts published. All public on Beehiiv.

But she had no clue which topics performed best. Which posts got shared most? Which patterns actually worked?

She could spend weeks manually reviewing. Or she could let AI do it.

Maya started typing her newsletter URL. Stopped.

"Wait, will it find enough? What if it only sees recent posts? What if it misses my best work?"

20 minutes second-guessing the URL. Zero progress.

The problem? Maya didn't trust AI could actually search and analyze her entire public archive.

But if she could just give AI her newsletter URL, it would search, identify top topics, and show her what patterns emerged.

Here's what Maya ran:

The public archive audit prompt:

Search this newsletter and identify:
- Top 10 most-discussed topics
- Writing patterns you detect  
- Most referenced/linked posts (if visible)

Then summarize what you found.

---

INPUT:
Your newsletter URL: [e.g., https://100mclub.beehiiv.com]

**Note:** If your posts aren't public, skip this step and go straight to Prompt 2 with manual copy/paste.

The AI sidekick searched her Beehiiv archive.

It scanned public posts. Identified topics. Detected patterns.

📥 Step 2: Maya had AI study her posts and generate titles

Maya had the topic: Writing productivity.

She had the top posts AI identified.

Now she needed two things:

Train AI on her voice and patterns
Get title ideas for new posts

She could manually read all 147 productivity posts. Take notes on patterns. Create a style guide. Then separately brainstorm 20 title ideas.

That would take days.

Or she could give AI the URLs of her top 5 posts and let it both study her patterns AND generate title ideas automatically.

Maya started gathering URLs.

"Should I pick the 5 most-linked posts? Or the 5 most recent? Or mix old and new?"

She picked 3 URLs. Started typing the prompt.

Wait. Is 3 enough for pattern recognition?

Added 2 more. Now 5 URLs ready.

Started typing: "Fetch these 5 posts and study my voice..."

Stopped. Wait, should she ask for titles in the same prompt? Or run it separately after?

If she runs separately, that's another prompt. Another waiting period.

But if she combines them, will AI do both tasks properly?

25 minutes second-guessing whether to combine the requests or split them.

The problem? Maya didn't know if AI could handle both pattern learning AND title generation in one go.

But if she could give AI 5 URLs and ask it to both study her patterns and generate 10 title ideas, she'd knock out two steps at once.

Here's what worked:

The pattern training + title generation prompt:

Fetch and study these 5 posts from my newsletter to learn my writing style:

[URL 1]
[URL 2]
[URL 3]
[URL 4]
[URL 5]

Study my: voice, structure, angles, transitions, opening hooks, closing CTAs.

Then, based on what you've learned, generate 10 title ideas for new posts about [Topic] that match my style.

Return:
1. Confirmation you've studied all 5 posts
2. 10 title options (with brief rationale for each)

---

INPUT:

**Required:**
5 URLs to your best posts on the same topic
Topic for new posts: [e.g., Writing productivity]

**Optional (skip if you've done context setup):**
Target audience: [e.g., Newsletter creators]

**Note:** If posts aren't public, paste the complete text of 3-5 posts here instead of URLs.

Maya pasted 5 URLs:

Topic: Writing productivity

The AI sidekick fetched all 5 posts.

Read the complete content. Analyzed opening hooks. Studied framework structures. Identified voice patterns. Noted transition styles.

Then it generated 10 title ideas:

"The 3-Minute Content Multiplication System" (Framework-based, matches your numbered approach)
"Why Your Writing Process Is Slowing You Down" (Problem-focused, conversational)
"How I Write 7 Posts Per Week Without Burning Out" (Personal achievement, specific metric)
"The Productivity Framework Top Creators Actually Use" (Authority + framework)
"Stop Starting From Scratch: The Template Method" (Command + solution)
"5 Writing Habits That Cut My Time in Half" (Numbered list, personal proof)
"The Daily Writing System I Wish I'd Found Sooner" (Personal regret hook)
"How to Turn 1 Hour of Writing Into 5 Posts" (Leverage ratio)
"Why Productive Writers Don't Start With Blank Docs" (Insight-driven)
"The 15-Minute Framework for Consistent Output" (Time + framework)

It confirmed: "I've analyzed all 5 posts. I understand your writing productivity approach: personal story openings, 3-step frameworks, conversational tone with strategic humor, clear completion moments. Pick any title above and I'll write the full post in your voice."

Maya scanned the list. #3 felt right. "How I Write 7 Posts Per Week Without Burning Out."

Specific metric. Personal achievement. Addresses the exhaustion pain point.

Completion moment: Maya had a trained AI that understood her patterns AND a title ready to go.

✍️ Step 3: Maya picked a title and generated the full post

Maya had the trained AI. It knew her writing patterns from 5 productivity posts. It had generated 10 title ideas.

Now she just needed to pick one and generate the full post.

She looked at the 10 titles again.

#1 looked good. But #3 felt more personal. Wait, #8 had a better leverage ratio.

Started second-guessing. "Should I pick the one with the clearest metric? Or the most curiosity-driven? Or the one that matches my recent posts?"

15 minutes comparing titles. Still hadn't picked one.

The problem? Maya was overthinking which title would perform best instead of just testing one.

But if she could just pick any title from the list and let AI generate the full post, she could test it live and iterate later.

Here's what she ran:

The post generation prompt:

Great. I'll go with title #[X]. Please write an 800-word newsletter post based on this headline, adopting my same writing style you just learned:

Title: [Copy the title from the list above]

Please write the complete post now.

---

INPUT:
[Use the trained AI from Prompt 2 in the same chat]

Selected title number: [e.g., 3]
OR
Custom title: [If you want to use your own instead]

Maya typed: "I'll go with title #3: 'How I Write 7 Posts Per Week Without Burning Out'"

The AI sidekick returned a complete 800-word post.

Opening hook about creator exhaustion. Personal story in her style (not Maya's exact story, but matching her voice). Framework breakdown showing 3 steps. Specific time metrics (15 minutes per post). Conversational tone with strategic humor. Clear completion moment at the end.

Maya read through it. Structure? Solid. Voice? Close to hers. Content? 7/10.

The formatting was off. No bolded subheads. Dense paragraphs. No bullet breaks. Transitions needed work.

But the skeleton was there. The ideas were there. The flow was there.

10 minutes of editing:

Break dense paragraphs (2-3 sentences max)
Add bold subheads every 3-4 paragraphs
Insert bullet points for lists
Smooth 3-4 awkward transitions
Add one personal story from her real experience

Done. Ready to publish.

Completion moment: Maya had a 7/10 post that took 10 minutes to finish instead of 5 hours to create from scratch.

🏆 Maya's results after 3 weeks

Before:

Posts per week: 1-2 (starting from scratch each time)
Time per post: 5 hours (research + writing + editing)
Mental energy: Exhausted by Wednesday
Archive usage: 0% (never touched old posts)

After:

Posts per week: 5-7 (multiplying existing content)
Time per post: 15 minutes (10 min editing + 5 min publishing)
Mental energy: Fresh all week
Archive usage: 100% (every post builds on proven patterns)

Her process now:

AI searches her newsletter URL and identifies top topics (60 seconds)
AI fetches 5 top posts, learns patterns, generates 10 title ideas (90 seconds)
Pick a title and AI generates full post (60 seconds)
Clean up formatting and add personal touches (10 minutes)
Publish

Total time: 15 minutes per post. Not 5 hours.

Her AI sidekick handles archive search, pattern analysis, title generation, and first draft in 3 minutes total. Not bad.

🧩 Your turn

Copy all 3 prompts into your AI sidekick. Run them in the same chat.

If your posts are public: Start with Prompt 1: Give your newsletter URL, AI audits your archive.

Then Prompt 2: Give 5 URLs to top posts, AI fetches them, learns your patterns, and generates 10 title ideas.

Then Prompt 3: Pick a title (or use your own), AI generates the full post.

If your posts aren't public: Skip Prompt 1. Just copy/paste 3-5 complete posts into Prompt 2. Same result.

Generation time: 3 minutes total. Time to publish: 15 minutes.

That's it, my fellow outliers!

Yours 'helping solopreneurs skip the hard way of doing things' Vijay peduru 🦸‍♂️