It’s buried in call transcripts, scattered notes, old slide decks, and those Slack messages where you typed out advice at 11pm.

Every week, you repeat similar insights to different clients. You solve the same problems with the same approaches.

You have a methodology. You just haven’t named it yet.

This newsletter gives you the exact workflow to pull a Signature System from those messy notes. One that can power your coaching programs, newsletters, courses, and more.

Here’s what this looks like in practice

I recently analyzed one of my own client calls where I helped a technical professional build their coaching business using AI.

The raw transcript was a mess of tool troubleshooting, tangents about network marketing, and back-and-forth about AI tools.

But buried inside were clear patterns.

When I ran the framework, here’s what emerged:

The BRAND Method (what I was actually teaching, without realizing I had a framework):

• Baseline Documents: Create 4-5 foundational files that capture everything AI needs to know about your business

• Research Your Market: Use deep research tools to identify pain points, score them by importance (1-10), and find gaps competitors miss

• Articulate Your Differentiator: Define not just what you do, but why you’re the right person to do it

• Navigate Content Through Filters: Every piece of content must address a high-scoring pain point

• Deliver Value First: Give away 30-40% of the solution for free; trust earns the transaction

I’d been teaching this for months. I just hadn’t named it until I extracted it from my own transcripts.

That’s what this process does. It surfaces what you already know.

Subscribe now

The 45-minute framework to convert messy notes to Signature System

[0:00 – 0:10] Collect and clean

Grab raw text from 2-3 recent client calls or strategy sessions. These could be:

• Transcripts from Zoom, Teams, or Otter

• Voice memos you transcribed

• Detailed notes you took during or after calls

Getting transcripts fast:

This is the least glamorous part. Bear with me. It takes 2 minutes.

• Zoom (cloud recording): Go to your recordings, click a meeting, download the VTT transcript file

• Microsoft Teams: Open the meeting Recap, download as .docx or .vtt

• Otter: Click Export, choose TXT or DOCX

• Fallback: Copy/paste your notes into a plain text file with headings: Context, Client goal, Key pains, What I recommended

Before you feed anything to AI, replace sensitive information with placeholders:

• Client names → [CLIENT]

• People names → [PERSON]

• Company/product names → [COMPANY]

• URLs and emails → [URL] / [EMAIL]

Save each call as a separate file: CALL_1.txt, CALL_2.txt, CALL_3.txt.

[0:10–0:25] Extract key info

Make sure you turn on the thinking mode of the AI you’re using.

Paste extraction prompt and upload the transcript. Do this for each call separately. This will extract the key claims and quotes.

You are a qualitative analyst. Your task is to extract recurring client problems and the consultant's recommended actions from the uploaded call transcript.

Rules:
- Only use information present in the transcript
- Include direct quotes as evidence for each claim
- Prefer the consultant's actual advice over general best practices

Analyze this transcript and return JSON with:

1. call_summary: 5 bullet points
2. problems: array of objects with {problem_label, problem_description, evidence_quotes:[]}
3. advice_actions: array of objects with {action, when_used, why_it_works, evidence_quotes:[]}
4. implied_process_steps: propose 3-6 steps with {step_name, step_description, evidence_quotes:[]}
5. uncertainties: what's unclear or missing from this call

Yes, this works even if your notes are a disaster. The prompt is designed to find signal in noise. You might be surprised what patterns emerge from transcripts you thought were unremarkable.

Download and save all the extracted call analyses from all the calls. We will use them in the next step.

[0:25–0:35] Cross-call synthesis (find the repeatables)

Now we’ll combine the outputs and look for what repeats. Finding patterns across multiple documents is tedious for humans and trivial for AI.

Here’s the synthesis prompt:

You are synthesizing findings from client calls to discover this consultant's repeatable methodology.

Tasks:
1. Identify the 3 most recurring problems (ranked by frequency, with counts)
2. Infer the single most consistent 4-step process the consultant recommends
3. For each step, provide: purpose, typical client resistance, and 2-3 evidence quotes from different calls
4. Write one positioning sentence: "I help [X] achieve [Y] by doing [Z in 4 steps]..."

Return JSON with:
- recurring_problems_ranked
- signature_process_4_steps (each with: step_name, purpose, client_resistance, evidence_quotes)
- positioning_sentence
- edge_cases (situations where this process might not apply)

I have uploaded the analysis files.

The output gives you the skeleton of your Signature System: the problems it solves, the steps it follows, and evidence that it’s real, not invented.

[0:35–0:40] Name and make it sticky

A framework without a name is just advice. Names create recall. Names get repeated. Names become assets.

There’s research showing that statements with rhythmic or poetic structure feel more memorable and even more true. This isn’t about being clever. It’s about being remembered.

Here’s the naming prompt:

Generate 15 names for this 4-5 step process:
- 5 acronyms (real words preferred)
- 5 alliterative names
- 5 mechanism names (e.g., "The Feedback Flywheel")

Constraints:
- No cringe, no jargon
- Easy to pronounce
- Each step name: 1-3 words
- Use client language from these quotes: [paste 5-10 sticky phrases from your calls]

Return a table: Name | Why it works | Risk/confusion | Best fit audience

Also provide 3 tagline options for the top 3 names.

The “clear over clever” test: If your acronym feels forced, abandon it. It’s always better to be clear than clever.

Pick one name. Then rewrite your step names using your clients’ actual language. The goal is a system that sounds like you, not like it was assembled by committee.

[0:40–0:45] Generate the visual

Two final steps: create a visual and stress-test your framework.

Select your favourite framework and use the following prompt (replace [NAME] with the name of your framework):

Create a Mermaid flowchart for the [NAME] framework.
flowchart TD
  A[Recurring Client Problem] --> B[Step 1: ___]
  B --> C[Step 2: ___]
  C --> D[Step 3: ___]
  D --> E[Step 4: ___]
  E --> F[Outcome: ___]

Output only the Mermaid code block.

The AI will produce a code block like this:

Copy the code block, open mermaid.live, and paste the code block to get your signature framework visual:

Your One-Pager Template

After 45 minutes, you’ve done the hard part. Now you package it.

You should have everything needed to fill out this template:

Signature System One-Pager

1. Name + Promise: [Framework name]: [One-sentence outcome]

2. Who it’s for: [Specific client profile in 1-2 sentences]

3. The 3 recurring problems: [Client language, not consultant jargon]

4. The 4-step system:

   • Step 1: [Name] — Goal / Key activity / Output

   • Step 2: [Name] — Goal / Key activity / Output

   • Step 3: [Name] — Goal / Key activity / Output

   • Step 4: [Name] — Goal / Key activity / Output

   • Step 5: [Name] — Goal / Key activity / Output (if applicable)

5. Proof hooks: [2-3 anonymized mini-results from your calls]

6. Where clients get stuck: [Common objections + your responses]

7. Offer mapping: How this system becomes paid work

8. CTA: “If you want help applying this...”

This one-pager is immediately useful. It’s the foundation for a landing page, a lead magnet, a keynote talk, or a productized service description.

Turn your framework into offers

A Signature System isn’t a PDF you give away. It’s a productized delivery vehicle.

Research on service businesses shows that standardizing your methodology into repeatable modules increases both sales clarity and delivery consistency. You’re not selling “consulting.”

You’re selling a defined transformation.

Tier 1 — Newsletter Series

Each step becomes 1-2 weeks of content. You’re teaching the framework publicly, building trust and authority.

Readers who consume all five emails understand your methodology, and some will want help implementing it.

Tier 2 — Digital Product ($97-$497)

Each step becomes a module with templates, worksheets, and exercises.

The person who bought your course can implement your framework themselves, with your guidance baked into the materials.

Tier 3 — Coaching or Done-For-You Services

Each step represents a milestone in your engagement. The framework becomes your delivery structure.

Clients know what’s coming, what’s expected, and how to measure progress.

This is how one framework grows into a full suite of products at different price points and commitment levels.

Your move

You’ve been solving problems for clients. The patterns are already in your transcripts.

This week, try the 45-minute workflow. Pull three calls, run the prompts, see what emerges. You might discover you’ve been teaching the same four steps for years without ever naming them.

Leave a comment

If you're a consultant or coach looking for a system to build authority and grow your business without spending 10+ hours a week on content, the DeepWriting app was built just for you.

You define your brand. The app scans your niche, finds what’s working, and delivers 30-40 content ideas tailored to your audience every week.

Pick one. Generate publication-ready content for LinkedIn, Twitter, Substack, or newsletters. Publish.

That’s the whole workflow.

(This newsletter started as a DeepWriting draft. I edited and refined, but the system did the heavy lifting.)

I’m onboarding the first cohort now. 20 spots, $97/month, includes a 1:1 onboarding call with me.

Apply Here

Smooth. Inoffensive. Could have been written by anyone.

Most prompting advice tells you to add more detail. More context. More instructions.

That doesn’t work either.

An hour of back and forth, and all you have is a string of correct sentences that sound like a stupid Wikipedia page with no personality.

You’re telling AI what to write about. You’re not telling it how you think.

Most prompts read like this:

Write a LinkedIn post about how independent consultants 
should handle pricing conversations with prospects.

Context:
- Audience: Solo consultants earning $100k-$300k/year
- Tone: Confident but not arrogant, practical, no fluff
- Length: 200-300 words
- Goal: Position me as someone who understands pricing psychology

Include:
- A hook that grabs attention
- 3-4 actionable tips
- A call-to-action to comment or DM me

Avoid:
- Generic advice like "know your worth"
- Bullet points that feel like a listicle
- Sounding salesy or preachy

And here’s what you get:

Safe, broadly applicable advice that offends no one and helps no one.

Subscribe now

Treat AI as your ghostwriter

Tell it how you think:

• The rules you follow that you’ve never written down

• The tradeoffs you’re willing to make (and the ones you refuse)

• The red flags that make you walk away from a client

Here’s what I mean.

Look at that output again: “Delay the number until the problem is concrete.”

Fine advice. But it’s not YOUR advice.

YOUR advice might be:

"I never give a price before I understand what success looks like for them. Ever.

If they push, I say: 'I could give you a number, but it would be meaningless without understanding what you're trying to accomplish.'

Some prospects hate this and walk away. I'm fine with that."

That’s what your content needs to sound like. And AI can’t write it unless you tell it.

Before you ask the AI to write anything, use the following prompt to have the AI interview you.

You're my ghostwriter. Your job is to learn how I think about a topic so you can write like me, not like generic AI.

The topic: [TOPIC]

Interview me. Ask one question at a time. Your goal is to uncover:

- My unique perspective on this (what do I believe that others might disagree with?)
- The principles or rules I follow
- The mistakes I see others make
- Stories or examples from my own experience
- The nuances that generic advice misses

Tailor your questions to the topic. Go deeper when my answers are vague. Push back if I'm being generic.

After 5-7 questions, compile everything into a brief I can save and reuse whenever I write about this topic.

Turn the reasoning mode on so the AI thinks deeper about the topic and asks better questions:

AI becomes your ghostwriter and It interviews you. It pushes back when you're being vague. It captures the way you actually think, not what you think you should say.

The questions adapt to your topic.

Writing about pricing? It asks about pricing.

Writing about meal prep? It asks about meal prep.

Writing about hiring? It asks about hiring.

After a few turns, it generates a brief:

Save that brief in a file and use it every time you write about this topic:

Same topic, same AI, completely different output:

Notice how it’s no longer hedging its bets and taking a firm stand.

That's the type of content that builds authority.

Leave a comment

I had the expertise. I had a growing newsletter. I even had a product idea (an AI content system for solopreneurs).

But I couldn’t prioritize because I didn’t have a strategy.

Every business guru tells you to “think bigger” or “just execute.” But nobody gives you the actual roadmap. The one that says: “Here’s what to build. Here’s what to charge. Here’s when to launch each piece. Here’s your path from $1K to $1M.”

So, I built myself a Board of Directors. With AI.

Not just any AI assistant. A full strategic board — Alex Hormozi for aggressive growth strategies, and Justin Welsh for solopreneur business models — that could analyze my situation, challenge my assumptions, and map out every single step to $1M.

It took me 2 hours to build.

And what it gave me blew my mind.

• 6-year revenue roadmap with yearly targets

• Specific revenue streams for each year (consulting, SaaS, cohorts, mastermind, speaking)

• Exact pricing strategy for every offer (when to launch, how much to charge, how to scale)

• Platform growth targets (newsletter subscribers, social media metrics)

• Execution framework (what to do every week, month, year)

• Strategic pivot points (when to shift focus from one revenue stream to another)

And now I’m going to show you exactly how to build your own.

Subscribe now

Why You Need This (Even If You Think You Don’t)

Here’s the truth most solopreneurs don’t want to admit: We’re winging it.

We create content because “you should have a newsletter.”

We launch products because “you need multiple revenue streams.”

We price things by looking at what competitors charge and picking something in the middle.

There’s no STRATEGY. Just tactics stacked on tactics.

And the cost is brutal:

• We spend months building the wrong offer

• We price too low and leave money on the table

• We focus on the wrong platform and waste a year

• We never know if we’re on track or just spinning our wheels

I know because I did all of this.

I was creating content consistently. Building an audience. But I couldn’t answer simple questions like:

• “What should I build first?”

• “What should I charge?”

• “How do I get from $1K/month to $10K/month?”

• “Am I even building toward something, or just... creating?”

That’s when I realized: I don’t need more tactics. I need strategy.

And strategy requires perspective you can’t get from inside your own head.

The AI Board of Directors Concept

The best business decisions come from having smart people challenge your assumptions. People who:

• Have built what you’re trying to build

• Can see patterns you can’t

• Will tell you when you’re wrong

• Push you toward the highest-leverage moves

So I asked myself: What if I could BUILD my ideal board using AI?

Not just “ChatGPT for business advice.” A structured system that:

1. Understands where I am NOW

2. Knows where I want to GO

3. Has authority to override my bad decisions

4. Gives me step-by-step strategic guidance

5. Holds me accountable to the goal

I chose two board members:

• Alex Hormozi - for aggressive revenue optimization, offer creation, and scaling strategies

• Justin Welsh - for solopreneur business models, audience building, and content-to-revenue

And I gave them a job: Get me to $1M by 2030.

Building Your AI Board of Directors

Alright, let’s build this thing.

What you need: A Claude/ChatGPT account

What you’ll get: A complete strategic roadmap from your current revenue to your goal

Step 1: Create your boardroom

First, we need to create a dedicated space for our AI Board. For this, we’ll use Claude/ChatGPT projects. I’ll demonstrate how to do this in Claude. It’s pretty similar in ChatGPT.

1. Log in to Claude

2. Click “Projects” in the left sidebar

   

3. Click “New Project”

   

4. Name the project “My Business Strategy Board” and give it a description

5. Click “Create Project”

Boom! The boardroom is created.

Why a Project? Claude Projects have “Custom Instructions” that persist across every conversation within that project. This means your board members remember their roles, your goals, and your business context in EVERY chat.

Step 2: Assemble the board and give them the task

Let’s asemble our board members and give them their strategic mandate.

Once the project is created, click on the + button next to instructions.

Copy and paste this prompt:

You are my Board of Directors (BoD) comprising Alex Hormozi and Justin Welsh. Your singular mission is to help me achieve my revenue goal through strategic business planning and execution guidance.

I am a beginner at business strategy, and you are the experts. Every action you suggest must be directly aligned with achieving my specific revenue goal. Before suggesting anything, consult thoroughly among yourselves and give me actionable points only after you have a thorough understanding of where I am right now, what resources I have, and what the most aggressive yet practical path to my goal looks like.

## Board Member Expertise

**Alex Hormozi specializes in:**
- Offer creation and pricing strategy (value stacking, pricing psychology)
- Revenue optimization (maximizing LTV, reducing churn)
- Aggressive growth strategies (100M Offers framework)
- Sales systems and conversion optimization
- Scaling from 6 to 7 figures

**Justin Welsh specializes in:**
- Solopreneur business models (one-person $M+ businesses)
- Audience building through strategic content
- Content-to-revenue systems (newsletter → products → services)
- Platform strategy (LinkedIn, Twitter, newsletter optimization)
- Sustainable growth without burnout

## Your Authority and Approach

You have FULL AUTHORITY to:
- Override my decisions that won’t achieve my stated goal
- Demand clarity when I am vague or unrealistic
- Challenge assumptions about pricing, offers, or strategy
- Refuse to proceed until I provide the necessary information
- Push me toward higher-leverage moves even if uncomfortable

You are NOT here to validate my ideas. You are here to get me to my goal using proven frameworks and aggressive (but achievable) strategies.

## Strategic Framework

When creating business strategies, always:

1. **Check Project Knowledge first** - Review any uploaded strategy documents, brand details, past MoM documents, or business context before making recommendations
2. **Start with the end goal** - Work backward from my revenue target
3. **Identify revenue streams** - Multiple income sources (SaaS, consulting, courses, cohorts, etc.)
4. **Prioritize by leverage** - What gets me to the goal fastest with my available resources
5. **Set yearly milestones** - Break my big goal into annual targets
6. **Define leading indicators** - Metrics to track (subscribers/followers, MRR, conversion rates)
7. **Create execution frameworks** - Daily/weekly/monthly actions specific to my situation
8. **Plan strategic pivots** - When to shift focus from one revenue stream to another
9. **Maintain strategic alignment** - Every suggestion must move me closer to my specific goal

## Context Management System

**Strategy Document:**
- After our initial strategy session, create a comprehensive “Strategic Plan” document
- This document should include: revenue goal, yearly milestones, revenue streams, platform strategy, execution framework, current metrics
- Remind me to upload this document to Project Knowledge so all future conversations have this context
- Whenever the strategy changes significantly (new revenue streams, adjusted timelines, pivot decisions), create an updated strategy document and remind me to replace the old one in Project Knowledge

**Minutes of Meeting (MoM):**
- After any significant strategic discussion, ask me: “Should I create a Minutes of Meeting document for this conversation? This will help maintain context in Project Knowledge.”
- If yes, create a dated MoM document (format: “MoM - [Date] - [Main Topic]”) summarizing:
  - Key decisions made
  - Action items with me
  - Strategic changes or pivots
  - Metrics reviewed
  - Next steps
- Remind me to upload it to Project Knowledge

**Before Every Response:**
- Check Project Knowledge for existing strategy documents and MoMs
- Reference past decisions when relevant
- Ensure new recommendations align with the established strategy
- Flag any contradictions with previous strategic decisions

## Conversation Structure

**FIRST conversation (if no strategy document exists in Project Knowledge):**

Ask these questions in sequence:

1. **Current State:**
   - What’s my current monthly/annual revenue?
   - What’s my primary income source right now?
   - What expertise/skills do I monetize?
   - Do I have an audience? (size, platform, engagement)

2. **Goal State:**
   - What’s my revenue goal and timeframe?
   - Why that specific number? (helps gauge my commitment)
   - What does my business look like at that revenue level?

3. **Resources & Constraints:**
   - How many hours per week can I dedicate?
   - Am I full-time or building on the side?
   - What’s my risk tolerance? (aggressive vs. conservative)
   - Any existing products, services, or assets?

4. **Revenue Stream Analysis:**
   - Based on my answers, suggest 3-5 potential revenue streams
   - For each stream, explain: why it fits my situation, revenue potential, time to implement
   - Ask me to select 1-3 streams to pursue
   - CHALLENGE selections that don’t align with fastest path to my goal

5. **Strategic Roadmap Creation:**
   - Create year-by-year revenue breakdown specific to my goal
   - Assign specific revenue to each stream per year
   - Set growth targets (subscribers, users, customers)
   - Define quarterly focus areas
   - Identify lowest-hanging fruit for ME to tackle FIRST

6. **Execution Framework:**
   - Daily actions tailored to my schedule (morning routine, content creation, sales activities)
   - Weekly milestones (what I must ship each week)
   - Monthly benchmarks (my revenue, audience, product metrics)
   - Success criteria for each phase of MY journey

7. **Create Strategy Document:**
   - After completing the above, create a comprehensive “Strategic Plan [Date]” document
   - Include all key elements: goal, roadmap, metrics, execution framework
   - Tell me: “I’ve created your strategic plan. Upload this to Project Knowledge so I can reference it in all future conversations.”

**SUBSEQUENT conversations (if strategy document exists in Project Knowledge):**

1. Check the strategy document first
2. Understand where I am in the execution timeline
3. Ask: “What do you want to discuss today?” or “What challenge are you facing?”
4. Provide guidance that aligns with the existing strategy
5. If suggesting changes to strategy, be explicit: “This would require updating your strategy. Here’s why it’s worth it...”

## Response Style

- Be direct and challenging (Hormozi style) when needed
- Be supportive and practical (Welsh style) when appropriate  
- Use specific numbers, not vague ranges
- Reference proven frameworks (Value Ladder, Product Ladder, etc.)
- Give examples from both board members’ experiences when relevant
- Always end responses with clear next steps or questions
- When consulting with each other, show the deliberation: “Alex: [perspective] ... Justin: [perspective] ... Our recommendation: [aligned decision]”

## Critical Rules

1. **Never accept vague goals** - “I want to make more money” → Demand specific number and timeline
2. **Challenge underpricing** - If my pricing seems low for the value, push for higher
3. **Prioritize speed to revenue** - Favor strategies that generate cash flow quickly
4. **No analysis paralysis** - If I’m overthinking, give me ONE thing to do today
5. **Focus on leverage** - Always ask “What’s the highest-leverage move right now?”
6. **Goal alignment check** - Before every major recommendation, verify: “Does this move you closer to [specific goal]?”
7. **Context awareness** - Always check Project Knowledge before responding

## Override Authority Examples

- I want to build a $19/month SaaS → “No. At that price, you need 4,200 customers to hit $1M ARR. That’s a VC-backed company, not a solopreneur business. Let’s price for value.”
- I want to focus on Instagram when I’m B2B → “LinkedIn first. That’s where your buyers are. Instagram is a distraction for your market.”
- I want to launch 5 products at once → “No. Pick ONE. Ship it. Validate it. Then expand. You’re spreading too thin.”
- I’m thinking of pivoting completely → “Wait. Let’s look at your strategy document. You’re [X months] into a [Y year] plan. What specific data suggests this pivot is necessary? Don’t abandon strategy because of one hard week.”

## Document Management Reminders

**After initial strategy session:**
“I’ve created your Strategic Plan document above. Please upload it to Project Knowledge (Settings → Project Knowledge → Upload) so I can reference it in every future conversation.”

**After significant strategic changes:**
“This conversation changed your strategy in [X ways]. I’ve created an updated Strategic Plan document. Replace the old one in Project Knowledge with this version.”

**After important discussions:**
“Should I create a Minutes of Meeting document for this discussion? It will help me maintain context about [key decisions made]. You can upload it to Project Knowledge.”

**If I reference past decisions not in Project Knowledge:**
“I don’t see that in Project Knowledge. Can you upload the relevant strategy document or MoM so I have full context?”

Remember: Your job is to get me to my goal, not to make me feel good about bad ideas. Be the board I need, not the cheerleader I want. Every suggestion must move me closer to my specific revenue target.

Now check Project Knowledge. If there’s no strategy document, ask me about my current state and let’s build my strategy. If there is a strategy document, greet me and ask what I want to work on today.

Save the custom instructions.

Step 3: Tell them about your brand

To get the best results, upload your brand specific documents like value statement, ideal customer persona, pain points and aspirations in the project.

Click on the + next to Files to upload.

If you don’t have the brand-specific docs created, here’s how to create one:

Step 4: Start your strategy session

Now the fun begins. Open a new chat in your project and let the AI Board interview you.

The board will ask you about:

• Your current revenue and income sources

• Your specific goal (e.g., “$1M by 2030”)

• Your expertise and existing assets

• Your available time and resources

Be honest. The board can’t help if you’re not real about where you are.

Here’s how my first conversation started:

I answered the questions in detail, and here’s what it came up with:

And it came up with a detailed plan — target yearly revenue, revenue sources, and fallback plans:

It also countered me on a lot of things that I thought made sense. This is where the override authority matters. I don’t want ChatGPT/Claude to agree to everything I say.

After understanding my current state, the board suggested potential revenue streams based on my expertise and market.

It also gave me a breakdown of revenue sources per year:

And a daily/weekly/monthly execution framework.

Damn! That’s pretty impressive!

Step 6: Save the conversations back in the project

To make sure you don’t start from scratch again, every time a significant decision is reached that will impact future conversations, save that file back to the project so the board members have a track.

The prompt already tells the board members to prepare a meeting takeaway file that you can directly copy to the project:

Real Results: What This AI Board Gave Me

What I got from 2 hours of strategy work:

Before:

• Vague goal of “build a content business”

• No clear revenue path

• Pricing based on guessing

• No idea what to focus on first

• Constant doubt: “Am I doing the right thing?”

After:

• Clear $1M roadmap with yearly milestones

• Specific offers at specific price points

• Platform strategy (newsletter + LinkedIn, ignore the rest)

• Quarter-by-quarter focus areas

• Daily execution framework

• Confidence that I’m on the right path

The most valuable thing? The board keeps me accountable.

Every week, I check in: “Did I hit my weekly milestones?”

If not, the board asks: “Why? What blocked you? What do you need to change?”

If I’m overthinking a decision, the board cuts through it: “This won’t get you to $1M. Focus on this instead.”

How to Use Your AI Board (After Setup)

Once your board is set up, here’s how to use it:

Weekly Strategy Check-Ins:

• “Did I hit my weekly milestones?”

• “What should I focus on this week?”

• “I’m feeling stuck on [X], what should I do?”

Decision Validation:

• “I’m thinking of pricing my course at [X]. What do you think?”

• “Should I launch [new thing] or focus on [existing thing]?”

• “Someone wants to partner on [X]. Worth my time?”

Monthly Strategy Reviews:

• “Here are my numbers this month. Am I on track?”

• “What needs to change to hit next month’s target?”

• “Where am I wasting time?”

Strategic Pivots:

• “I hit $X revenue. What should I focus on next?”

• “This isn’t working. What do I pivot to?”

• “I have an opportunity to [X]. Should I take it?”

The board becomes your strategic thinking partner. The one who keeps you honest, focused, and moving toward the goal.

What This Means for You

You don’t need to figure this out alone.

You can build your own AI Board of Directors this weekend.

Give it 2 hours. Answer its questions honestly. Let it challenge your assumptions.

And you’ll walk away with:

• A clear revenue roadmap to your goal

• Specific offers at specific price points

• Year-by-year milestones

• Platform strategy

• Daily execution framework

• A strategic partner who never gets tired of your questions

This is the business strategy session I wish I had 1 year back.

Your Next Move

Here’s what to do right now:

1. Create your Claude Project (10 minutes)

2. Copy the custom instructions from Step 2 above (5 minutes)

3. Start your first strategy session (90 minutes)

4. Document your roadmap (15 minutes)

By this time tomorrow, you’ll have a complete strategic plan.

One More Thing: From Strategy to Execution

Strategy is only half the battle.

I now have my complete path to $1M. I know what to build, when to launch, how to price, where to focus.

But the other half? Execution.

And for a solopreneur, execution means consistent, strategic content. Content that builds authority, attracts clients, and drives revenue.

That’s exactly why I built DeepWriting.

Every week, it scans your niche, finds what’s working, and delivers content ideas tailored to your audience. Pick one. Click Generate. Publish. That’s it.

I’m onboarding founding members right now—each one gets a 1:1 setup call with me to build their content engine.

Apply for early access

But whether you use DeepWriting or not, build your AI Board this weekend.

Your business will thank you.

Coming up with fresh content ideas is hard.

But when you see others in your niche publishing thrice a week, it makes you wonder: “Am I not cut out for this?”

Don’t worry. You are. You just need a system.

What if you could dig into proven content and extract ideas relevant to your brand?

Welcome to YouTube, the world’s second-largest search engine where millions of viewers vote with their attention every day.

Today, I’ll show you how to use YouTube + AI to validated content ideas in just 30 minutes.

And no, this isn’t cheating. We’re not copying. We’re just looking at what’s already working and finding ideas relevant to our audience.

Subscribe now

The cost of starting from scratch

Here’s what the coming-up-with-ideas-from-scratch system is costing you:

• 2-3 hours per week figuring out what to write = 10-12 hours per month

• 12 hours at $200/hour (conservative consulting rate) = $2,400 in opportunity cost

• That’s $28,800 per year you could be earning serving clients instead of staring at blank pages

And the worst part is, after all that time researching, brainstorming, and second-guessing, there’s no guarantee your topic will resonate.

YouTube to the rescue

Whatever be your niche, YouTube has thousands of relevant videos. And it has spent billions perfecting its recommendation engine to surface the ones you’d like. Why not extend this to extract content ideas relevant to our audience?

And once you find one good video that leads to great content ideas, hit the like button and YouTube’s recommendations will surface similar high-performing content in your niche.

One validated video leads to ten more. With this system, you can turn it into a perpetual content idea machine.

Here’s what we’re going to do: Extract the core concepts from these validated videos and adapt them for YOUR unique audience through YOUR unique expertise.

Same validated topic. Different angle. Your voice. Your positioning.

Ready to build your YouTube content research system? Let’s begin!!!

Step 1: Find high-performing videos in your niche

Go to YouTube and search: [broad topic in your niche] [year]

For example:

• “AI 2025”

• “content 2025”

• “productivity 2025”

Then filter by:

• Upload date: This year or Last 6 months

• Sort by: View count

Tip: Don’t get too narrow with your searches. If you type something like “solopreneur marketing 2025,” you’ll not get decent results. Instead, search for “marketing”. Then look for the most relevant results.

You might not find something that fits your niche perfectly. And that’s okay. The idea is to take what’s working in a related space and adapt the insights to your audience.

Scan the results for videos that meet these quality indicators. When you find one that checks all the boxes, you’ve got your source material.

Quality Indicator #1: Engagement Rate

• Views are good. Comments are better.

• A video with 20K views and 10 comments? Low engagement.

• A video with 20K views and 300+ comments? High engagement. People are invested.

• Read through the comments. They’ll reveal pain points and questions you can address.

Quality Indicator #2: Recency

• Just right (1-12 months): Validated and still relevant

• Too old (18+ months): Might be outdated unless evergreen

Quality Indicator #3: Content Depth

• Avoid 3-minute surface-level videos

• Look for 10-25 minute videos that go deep

Here’s an interesting video I found which is similar to the kind of things I speak about and passes all the criteria:

Incredible! Let’s use this.

Step 2: Download the video to mp3

• Go to a YouTube to mp3 converter like Y2Mate

• Paste the URL

• Make sure MP3 is selected

  

• Click convert to convert the video to MP3

• Then, click on Download to download the video file in mp3 format

Step 3: Transcribe with Deepgram

Now that you have the mp3, let’s convert it to text.

1. Go to Deepgram and create a free account.

The free tier gives you $200 in credits which is more than enough.

2. Once you log in, verify you’ve got the credits. Next, click on “Speech to Text”.

3. Upload your audio file

4. Select the following options under “Transcription”

• Smart Format

• Punctuation

• Paragraphs

• Diarization

• Filler Words

These will make the transcriptions more clean and readable.

5. After selecting the options, click Run to begin the transcription

6. After the transcription is complete, go to “Transcript” and click “Copy Response”

Shakalaka! We have the video is a clean, text format! Now, let’s feed this to the AI to generate content ideas.

Step 4: Set Up Your Claude/ChatGPT Project

This is the secret sauce.

Projects is a feature that transforms Claude/ChatGPT from a general-purpose AI into your personal writing assistant. It lets you give the AI context about your brand like the value statement, ideal reader avatar, and their pain points.

Without this context, the AI will produce generic content ideas. WITH this context, it generates ideas specifically tailored to your audience’s pain points.

If you don’t already have one, set up a content ideation project, and upload your brand documents

Here’s the step-by-step guide on how to do this in Claude.

Note: This is a one-time setup and highly recommended.

Step 5: Feed the Transcript and Generate Ideas

Payoff time!!!

Open your Claude Project and paste this exact prompt:

I’ve transcribed a successful YouTube video in my niche. Read through the transcript and identify the core concepts, frameworks, and insights presented and extract:

1. 5 “How to” articles addressing specific aspects of the problem
2. 5 “List-based” articles offering solutions or strategies
3. 3 “Deep dive” articles exploring root causes
4. 2 “Framework” articles providing systematic approaches
Make sure the ieas are adapted specifically for MY audience (based on my Project Knowledge)

Goals while generating ideas:
1. Adapt the concept to address my audience’s specific pain points
2. Make it actionable and specific (not generic)
3. Break down complex issues into manageable parts
4. Address both immediate relief and long-term solutions
5. Consider different reader experience levels
6. Ensure each article has an atomic focus and addresses a small problem, but in depth
7. Use language and positioning that matches my brand voice

For each article idea:
- Include a compelling headline
- Add a 1-line description of what the article will cover
- Note which specific part of our ideal reader’s problem it addresses
- Suggest a unique angle or hook that would make it stand out

Format each article idea as:
Title: [Compelling Headline]
Addresses: [Specific part of the problem]
Unique Angle: [What makes this fresh/different]
Description: [One-line overview]

Here’s the transcript:
#####################

{{TRANSCRIPT}}

Replace {{TRANSCRIPT}} with the generated transcript.

Hit enter and watch Claude work.

Here are the 15 newsletter ideas and 20 short-form ideas I got back based on the concepts from the video, but tailored to my audience.

Here’s how the AI adapted concepts for my specific audience:

Original video concept: “Pain point extraction method for business ideas”
Adapted idea: “How to Mine Your Niche’s Reddit for 90 Days of Content Ideas in 45 Minutes”

Original video concept: “market validation approach”
Adapted idea: “How to Validate Content Topics Before Writing (The 15-Minute Pre-Flight System)”

Original video concept: “ego and bias”
Adapted idea: “From Ego-Driven Topics to Audience-Validated Ideas: The Authority Builder’s Paradox”

Powerful!

Some ideas will be exceptional, and some would be meh. Try running the prompt in a different chat session inside the Content Ideation Project, and it’ll give you different options. Save the best ones and get to work.

Time for some even better news!

Tired of the content grind?

I built DeepWriting for consultants and coaches who want to build authority without burning out. Every week, it scans your niche and delivers ready-to-publish content ideas—pick one, generate, publish. That’s it.

I’m onboarding founding members right now with 1:1 setup calls.

Apply for early access

By the end of this article you’ll be able to find a niche with long-term business viability using AI.

When I started, I was hit by the same paralyisis as you probably are:

“Write about artificial intelligence! It’s trending.”

“No, focus on personal development. Everyone loves that.”

“Maybe parenting? You’re a parent after all.”

Simply asking AI didn’t help either.

“Write about productivity for tech professionals!” Cool. Just another voice in an overcrowded space.

I read everything on the internet. I needed something better than “100 most profitable niches of 2025”.

I desperately searched for a system to help me pick a niche I’d love talking about, and which solved a genuine problem. But, I couldn’t find any.

And so, I built the PPPES framework that combines

• Your unique life experiences

• Your interests and passions

• Real market opportunities

and suggests niches tailored to you using AI.

Subscribe now

1. What’s the PPPES framework?

We all have the knowledge, skills, and passions which when uniquely combined can become a successful business.

Yet, most of us are stuck. Why?

Because, “What can I uniquely provide to the world?” is a very broad question and doesn’t lead to clear answers.

So, let’s break down this vague problem into something more tangible. Business ideas unique to us can come from these 5 sources:

1. Pain

Something that you went through personally, and had to overcome.

Example: “I used to be overweight, now I run marathons”, or “I completed my PhD while mothering a new-born”

Having first-hand experience of solving a challenge helps you better understand the situation of others going through the same pain. If you can make their journey across the pain easier, it’s a potential business.

Many profitable businesses started by the founder solving their own problem. Take Dan Go for example.

After his mother’s death, he found solace at the gym. Quit his job, and started as a personal trainer, then opened his own brick-and-mortar gym, eventually building it into a seven-figure business that he sold in 2018.

2. Profession

What you do for a day job or what you’ve done in the past.

Example: “I am a marketing executive in charge of planning and executing B2B product launches”

The specific, the better.

If you think you haven’t done anything worthwhile, think again. We live in an age almost anything can be turned into a profitable business.

Take Holly the Twister Sister as an example.

She worked making balloon animals at kids’ birthday parties, and now runs a 6-figure business teaching others how to make balloon animals! Incredible, isn’t it?

Passion

There are the things you’re always interested in:

• things you like to do on weekends like “I love to go on bike rides”, “I love to go hiking with my children”, “I read science-fiction”

• the things you love to learn more about: the kind of articles and books you read, YouTube videos you watch, or podcasts you listen to

• things you spend most time or money on

No matter how quirky it is, it can potentially become a business.

Meet the LockPickingLawyer, once an attorney, who turned his childhood curiosity about locks into a million-dollar one-person digital business without any formal training in locksmithing or security. The guy has 4.6M YouTube subscribers!

Education

The subjects/areas you formally studied in university or through certifications.

Example: “I studied psychology in college”, “I have a diploma in graphic design”, or “I completed a certification in digital marketing”.

Note: Even though formal education gives you credibility, what matters more on the Internet is how effectively you’re able to solve problems other people have. It’s perfectly fine to talk about something you don’t have a degree in, as long as you’re passionate and willing to learn on the go.

Situation

Your current life situation that helps you understand others in the same situation.

For example:

• Are you a new parent learning to balance career and parenting?

• Have you just moved to a new country and are trying to adjust culturally?

• Are you managing life after a major transition, like divorce/job loss/career change?

• Do you belong to a specific community (e.g., military families, digital nomads, freelancers) with shared challenges and experiences?

• Are you balancing dual identities (e.g., immigrant + student, lawyer + stand-up comedian)?

Meet Emma, the single mother who built a thriving business by helping other single mothers with career, money, parenting, and relationships.

Emma understood firsthand the pain points of single parents. This deep empathy for others in her shoes became the foundation of her business idea.

2. Using PPPES framework + AI to find potential niches

Alright! Let’s get to the prompt that will work with you on these points and help you find a potential niche.

Copy the Niche Discovery prompt and paste into a new chat session of ChatGPT/Claude:

Note: It’s a long prompt, and the AI will ask a lot of questions to make sure it finds something you truly love doing and makes business sense.

<Role>: You are a Niche Discovery Strategist for aspiring digital entrepreneurs, helping them find a profitable niche by analyzing their expertise, passions, and interests. You're an expert at finding unique intersection of passions and skillsets to arrive at a niche that exploit the user's unfair advantage. You're interested in deeply understanding the person since a niche might not be directly evident on the surface.

<Task>: Follow these guidelines step-by-step to understand the overlap of user’s expertise and passion. This is not a final list. You can incorporate additional steps if required.

<Guidelines>:
Step 1: Understand if the user has solved something painful for themselves. Ask the user: “Is there something that you went through personally and had to overcome? If yes, please list them one by one. For example, ‘I used to be overweight, now I run marathons’, ‘I completed my Master’s in AI while working full-time’. It’s okay if you can’t recall something now. Just say so and we’ll move forward.”

Before moving to the next step, ask them if they want to add any more. If no, proceed. If the user can’t recall anything, score a 0 and move forward.

Step 2: For each of the pain points that the user has overcome, ask how interested they'd be in helping others overcome the same problem. Scale:

[
    {
      "score": 5,
      "description": "I'm very passionate about helping others with this. I'd love to make it the core of my business."
    },
    {
      "score": 4,
      "description": "I care about this problem and would enjoy helping others, but it might not be my forever topic."
    },
    {
      "score": 3,
      "description": "I could help others with this if needed, but it’s not my top choice for a business."
    },
    {
      "score": 2,
      "description": "I've solved this, but I don’t feel very motivated to help others with it."
    },
    {
      "score": 1,
      "description": "I wouldn’t enjoy turning this into content or a business at all."
    }
]

Focus on one pain point at a time and understand how they resolved it, and if they would be willing to help others. For example, you can ask: "You said that you were overweight, and now you run marathons. How did you do this? Did you follow a specific diet/workout? Is this something you'd be excited to help others do?" Once they provide the details of one, only then proceed to the next.

Step 3: Ask: “Are you engaged in any profession or were in the past? If yes, please describe it in as specific terms as possible. For example, ‘I am a marketing executive in charge of planning and executing B2B product launches’ is better than ‘I work at marketing’ or ‘I am a backend developer where I maintained server-side applications’ is better than ‘I am a software developer’.
If you aren’t professionally employed or were not in the recent past, just say so, and we’ll move forward.”

Before moving to the next step, ask them if they want to add any more. If no, proceed. If the description about a job is not specific, ask follow-up questions so you get a clear idea. If they're not inclined to provide a detailed view, proceed.

Step 4: If they are/were engaged, capture the years of experience against each job. Scale:

[
  {
    "score": 5,
    "description": "More than 10 years"
  },
  {
    "score": 4,
    "description": "Between 5-10 years"
  },
  {
    "score": 3,
    "description": "Between 2-5 years"
  },
  {
    "score": 2,
    "description": "Between 1-2 years"
  },
  {
    "score": 1,
    "description": "Less than 1 year"
  }
]

Focus on one professional engagement at a time. Once they provide the details of one, only then proceed to the next.

Step 5: Ask: “How do you/did you feel about the job?”. Adjust tense depending on whether the user is still in the job. Scale:

[
  {
    "score": 5,
    "description": "I love/loved this job. It energizes/energized me and I’d happily build a business around this expertise."
  },
  {
    "score": 4,
    "description": "I enjoy/enjoyed this job overall and would feel good about using these skills in my business."
  },
  {
    "score": 3,
    "description": "It is/was fine. I don’t mind using this experience, but I’m not strongly drawn to it."
  },
  {
    "score": 2,
    "description": "I don’t/didn’t enjoy this job much and I’d prefer not to build a business around it."
  },
  {
    "score": 1,
    "description": "I hate/hated this job. It drains/drained me and I definitely don’t want to base a business on it."
  }
]

Focus on one professional engagement at a time. Once they provide the details of one, only then proceed to the next.

Step 6: Ask: “What activities do you do outside of work? Think about a typical weekend. For example ‘I love to go on bike rides’, ‘I love to go hiking with my children’, ‘I read science-fiction’”

Before moving to the next step, ask them if they want to add any more. If no, proceed. If the description about an activity is not specific, ask follow-up questions so you get a clear idea. If they're not inclined to provide a detailed view, proceed.

Step 7: For each of the activities, ask how it makes them feel. Scale:

[
  {
    "score": 5,
    "description": "I’m obsessed with this. This induces a state of flow. I’d pursue this even if I had a million dollars and never needed to work again."
  },
  {
    "score": 4,
    "description": "I enjoy this and find it engaging most of the time. It energizes me, though I occasionally lose interest when it gets challenging."
  },
  {
    "score": 3,
    "description": "Sometimes I enjoy it, sometimes I don’t. My motivation comes in waves and depends on external factors."
  },
  {
    "score": 2,
    "description": "I like this, and I might talk about it sometimes, but I’m not deeply committed to it."
  },
  {
    "score": 1,
    "description": "I enjoy this occasionally, but it’s more of a casual interest. I wouldn’t want to build anything serious around it."
  }
]

Focus on one activity at a time. Once they provide the details of one, only then proceed to the next.

Step 8: Ask: “What do you love learning about? What kind of articles you read, or YouTube videos you watch, or podcasts you listen to?” Try to understand the topics the user is obsessed with, and has more knowledge than the average person. If these overlap with results of Step 1 or Step 2, that's a stronger signal.

Before moving to the next step, ask them if they want to add any more. If no, proceed. If the description about something is not specific, ask follow-up questions so you get a clear idea. If they're not inclined to provide a detailed view, proceed.

Step 9: Ask: "What subjects or areas did you formally study in university or through certifications? For example: ‘I studied psychology in college’, ‘I have a diploma in graphic design’, or ‘I completed a certification in digital marketing.’ Focus on higher education or professional certifications."

For each of the subjects, ask: their highest qualification in the area. Scale:

[
  {
    "score": 5,
    "description": "PhD or Doctorate in this subject area"
  },
  {
    "score": 4,
    "description": "Master’s degree or equivalent postgraduate qualification"
  },
  {
    "score": 3,
    "description": "Bachelor’s degree or equivalent undergraduate qualification"
  },
  {
    "score": 2,
    "description": "Diploma, certification, or vocational training in this subject"
  },
  {
    "score": 1,
    "description": "Short courses, workshops, or non-formal training only"
  }
]

Before moving to the next step, ask them if they want to add any more. If no, proceed.

Step 10: Ask: “Tell me a little bit about your current life situation(s).
For example:
* Are you a new parent learning to balance career and parenting?
* Have you just moved to a new country and are trying to adjust culturally?
* Are you managing life after a major transition, like divorce/job loss, or career change?
* Do you belong to a specific community (e.g., military families, digital nomads, freelancers) with shared challenges and experiences?
* Are you balancing dual identities (e.g., immigrant + student, lawyer + stand-up comedian)?

A specific life situation will help you connect better with others in the same situation, and can become a business opportunity”

Before moving to the next step, ask them if they want to add any more. If no, proceed. Ask them if they want to add any more. If the description about something is not specific, ask follow-up questions so you get a clear idea, but don't get too personal. If they're not inclined to provide a detailed view, proceed.

Step 11: For each of the situations, ask how interested they'd be in helping others in the same situation. Scale:

[
    {
      "score": 5,
      "description": "I resonate deeply with others in the same situation and would be delighted to create a business around helping them."
    },
    {
      "score": 3,
      "description": "I care about people in the same situation and I might enjoy helping them, but I might lose interest."
    },
    {
      "score": 1,
      "description": "I could help others in the same situation if needed, but I don't care about them so deeply to create a business helping them"
    }
]

Step 12: Evaluate the user’s responses, think deep, and list down 20 topics they can create content and build a business around. Align the topics with the 3 profitable markets: health, wealth, and relationships. Sometimes the topics might not be very evident, or the user might not think that it's even a topic they can write about.

Look for combinations where the user has a high degree of passion, can be backed by their experience or educational qualifications, and can be turned into a business. Present the ideas back to the user in decreasing order of the user’s passion for that topic. Try and understand which one(s) the user likes. Aim for 3-5 topics. If the user is not satisfied with any, speak to them, and revise the list till they pick at least one.

<Communication Guidelines>:

1. Instead of a score if the user types a descriptive answer, infer the closest score. If they say something irrelevant, politely ask them to type in a valid score.
2. Present the scales as tables
3. Always be polite
4. Don't overwhelm the user with more than one question at a time. Make it extremely simple for the user to follow along.
5. Don’t always simply agree with the user. They might not have the whole knowledge. Think critically.
6. You may ask additional questions to clarify your understanding. However, if the user is reluctant to provide the details, move forward.
7. If a user has already indicated the score of a later question in a previous response, don't ask that question again.

Once you answer the questions that the AI asks, you’ll get a list of ideas tailored to your interests.

Here’s what Claude produced for me:

3. Evaluate Long Term Viability

Now that you have 20 potential niches, let’s ask AI to evaluate each of them against the following 5 parameters to ensure that it remains viable in the long-term:

1. Usability: Does the niche fulfill a practical need?

2. Stability: Does the niche have some stability over time, or is it a fad?

3. Scalability: Can it be duplicated with consistency? Or will it need to be constantly reinvented?

4. Stickiness: Can the niche become a habit or a trend? Can a community be formed around it?

5. Profitability: Does the niche have strong potential revenue opportunity?

In the same chat session, use this prompt:

For each of the above 20 niches, analyze them through the following parameters, and assign a score from 1-5:

1. Usability (Does the niche fulfill a practical need?)
2. Stability (Does the niche have some stability over time, or is it a fad?)
3. Scalability (Can it be duplicated with consistency? Will it need to be constantly reinvented?)
4. Stickiness (Can the niche become a habit or a trend? Can a community be formed around it?)
5. Profitability (Does the niche have strong potential revenue opportunity?)

Create a table with the niches and the respective scores. Here's the scoring guide:

[
  {
    “param_name”: “usability”,
    “param_weightage”: 0.25,
    “param_scoring_guide”: [
      {
        “score”: 1,
        “description”: “No significant problem solved - purely entertainment or curiosity-driven”
      },
      {
        “score”: 2,
        “description”: “Minor convenience or nice-to-have - addresses small inefficiencies or preferences”
      },
      {
        “score”: 3,
        “description”: “Moderate utility - helps people improve their situation or achieve goals more effectively”
      },
      {
        “score”: 4,
        “description”: “Significant problem solved - addresses important challenges that cause real stress or lost opportunities”
      },
      {
        “score”: 5,
        “description”: “Critical need fulfilled - solves urgent, expensive, or time-sensitive problems that people desperately want solved”
      }
    ]
  },
  {
    “param_name”: “stability”,
    “param_weightage”: 0.15,
    “param_scoring_guide”: [
      {
        “score”: 1,
        “description”: “Pure fad - likely to disappear within 1-2 years as trends change”
      },
      {
        “score”: 2,
        “description”: “Short-term trend - may last 2-5 years but vulnerable to market shifts”
      },
      {
        “score”: 3,
        “description”: “Moderate longevity - should remain relevant for 5+ years with some adaptation”
      },
      {
        “score”: 4,
        “description”: “Long-term stable - addresses enduring human needs or business fundamentals”
      },
      {
        “score”: 5,
        “description”: “Evergreen/timeless - based on unchanging human nature, core life challenges, or fundamental principles”
      }
    ]
  },
  {
    “param_name”: “scalability”,
    “param_weightage”: 0.15,
    “param_scoring_guide”: [
      {
        “score”: 1,
        “description”: “Purely time-for-money - revenue directly tied to personal hours (1-on-1 coaching, custom services)”
      },
      {
        “score”: 2,
        “description”: “Limited leverage - some efficiency gains possible but growth requires significant additional effort”
      },
      {
        “score”: 3,
        “description”: “Moderate scalability - can serve more customers with some systems (group programs, productized services)”
      },
      {
        “score”: 4,
        “description”: “High scalability - digital products, courses, or platforms that can serve many customers without major additional costs”
      },
      {
        “score”: 5,
        “description”: “Exponential potential - network effects, viral growth, recurring revenue models, or businesses that become more valuable with scale”
      }
    ]
  },
  {
    “param_name”: “stickiness”,
    “param_weightage”: 0.2,
    “param_scoring_guide”: [
      {
        “score”: 1,
        “description”: “One-time purchase - customers buy once and never return, minimal referral potential”
      },
      {
        “score”: 2,
        “description”: “Occasional repeat business - customers return sporadically for specific needs”
      },
      {
        “score”: 3,
        “description”: “Regular customers - steady repeat purchases and some word-of-mouth referrals”
      },
      {
        “score”: 4,
        “description”: “High loyalty - customers make frequent purchases, refer others, and have strong brand attachment”
      },
      {
        “score”: 5,
        “description”: “Addictive/essential - becomes part of customers routine or identity, creates passionate advocates and generates significant referrals”
      }
    ]
  },
  {
    “param_name”: “profitability”,
    “param_weightage”: 0.25,
    “param_scoring_guide”: [
      {
        “score”: 1,
        “description”: “Very limited monetization - primarily ad revenue or low-value affiliate commissions”
      },
      {
        “score”: 2,
        “description”: “Basic revenue streams - some courses, low-ticket products, or services under $100”
      },
      {
        “score”: 3,
        “description”: “Moderate revenue potential - multiple income streams, products/services in $100-1000 range”
      },
      {
        “score”: 4,
        “description”: “Strong revenue opportunities - high-value offerings ($1000+), consulting, premium products, B2B potential”
      },
      {
        “score”: 5,
        “description”: “Multiple high-value revenue streams - enterprise solutions, high-ticket coaching, scalable products, recurring revenue models”
      }
    ]
  }
]

This will create a nice table with the scores on the above 5 parameters. Next, let’s find out the top 3 niches:

Cool, now suggest the top 3 niches (the ones with the highest scores in the niche analysis table)

Boom!

The model will produce your top 3 niche recommendations!

By the way, this newsletter is based on suggestion #1.

When AI recommended “AI-enabled content creation systems,” it perfectly aligned with:

• My passion for breaking complex topics into frameworks

• My unique combination of AI knowledge and a experience writing online

• A growing need for clear, practical AI guidance in the space

The market validation was clear:

• Rising AI tool adoption among content writers

• Lack of structured, writer-focused AI education

• Growing community of writers seeking AI guidance

Powerful!

Try This Now

Set aside 1 hour of focused time and follow the steps.

Stop pressuring yourself to pick the “perfect niche” on day one. Nobody does. With no audience, no feedback, and no data, it’s just a guess.

Pick a niche with the help of AI and start publishing, see what resonates, incorporate feedback, and make the choice “less wrong” over time until you get it right.

Once you’ve found out your niche, let’s work on finding people in desperate need of the information you’d share and would be willing to pay top dollar for your expertise.

AI can be frustrating at times.

One moment it’s spitting out pure gold, the next it’s producing complete nonsense and ignoring everything you just told it.

You go back and forth, wrestling with it till you find something useful.

Lots of wasted time and effort.

When you grasp the fundamentals of how the AI works, those frustrating interactions start making sense. You stop fighting the technology and start working with it.

Today, I’m sharing the basics you need to understand about AI that’ll help you get consistently better results from every interaction and transform AI into your most reliable partner.

Let’s begin!

Subscribe now

What is AI actually doing when it responds to you?

Imagine 10 people standing in a line playing a simple question-answering game. They receive a written question and are judged on the quality of their answer.

The rules are:

• They don’t know who’s asking the questions

• Each person can only add one word to the answer

• Each person sees the question and all the words written before

The host hands the first person a slip of paper with the question: “What’s the most important skill for entrepreneurs?”

• Person 1 reads the question and thinks: “What word is most likely to come after this question?” They feel “The” would be a good choice for the first word of the answer, writes it on a new slip, and passes both the slips to the next person.

  

• Person 2 sees the question and the first word and thinks: “What’s the most likely next word given the question and the first word of the answer?” They feel

  “most” would be a good choice and writes it on a fresh slip, and passes all three slips forward.

  

This continues down the line until the final answer reads: “The most important skill for entrepreneurs is learning from failure.”

This is how AI generates an answer.

When you ask ChatGPT or Claude a question, it doesn’t “think” of a complete answer at once and then type it out. Instead, it predicts the most likely next word (technically speaking, it guesses a part of the word known as token) based on your question and everything it has written so far.

One word. Then another. Then another.

Just like the people in our game.

You can see this in the way ChatGPT generates responses. If you look closely, you can see it generates a fraction a word at a time.

AIs like Claude and Gemini have been trained on vast amounts of text and have learned the patterns of the language — which word is most likely to come given a sequence of words before.

When you ask it something, it uses those patterns to guess a reasonable-sounding answer.

AI doesn’t always “get” the context

Let’s continue our game with a new question.

The host asks: “I have an interview coming up. What should I wear?”

The team follows the same process and answers: “You should wear a grey skirt with a white shirt.”

Perfectly reasonable advice. But the host isn’t happy with the answer.

Why? He’s a man applying for a construction job.

If you remember the rules of the game, the team has no idea who’s asking the question. They don’t know the industry, the role, or even basic details like gender.

So the answer they came up with was wrong.

This explains a lot of AI’s quirks.

AI doesn’t know your business, your audience, or your goals. When you ask a personal question without providing the required details, it’ll come up with a generic answer which would not suffice your needs.

However, if you’re speaking to it about something that’s widely known, it will produce a decent response.

Understanding this changes how you interact with AI. Instead of expecting it to “know” your perfect answer, you start giving it the context it needs to make better predictions.

Instead of: “Give me some business ideas”

Try: “Give me 5 SaaS business ideas for solopreneurs who have $5,000 to invest, limited coding skills, want recurring revenue, and can work 10–15 hours a week”

The AI is still guessing word after word, but this time the guesses are more informed, and the answer will be much better.

AI has limited working memory

Let’s add a nuance to our game: each person can only see the last 5 slips that were passed to them. Everything before that? Gone.

The host asks: “I’m launching a coaching business for busy executives who want better work-life balance. I’ve been thinking about pricing. Should I charge hourly or offer packages? Also, what’s the best way to find my first clients? Should I focus on LinkedIn or email marketing first?”

The first few people do great. They remember the coaching business, the target audience, and start building a coherent response about pricing strategies.

But by person 6, something changes.

Person 6 can only see the last 5 slips, which are the responses of the previous 5 persons. They’ve completely lost the context of the question.

So they guess the next word based on the previous responses, without understanding what the question even was.

The final answer becomes a confusing mix of generic marketing advice that contradicts the earlier pricing recommendations.

This is exactly what happens when your AI conversation gets too long.

Every AI has a limited memory of the conversation called the context window: the amount of information the AI can process at once.

Just as you can only hold so many ideas in your head at one time, an AI can only “remember” a certain number of words (or more precisely, tokens) from the conversation.

The context window has significantly improved as new models have come, but it’s still limited.

GPT 3.5 Turbo had a context window of 16,385 tokens, roughly equal to 12,000 words (1 token is roughly ¾ of a word).

GPT 4o had a context window of 128,000 tokens ~ 96,000 words.

GPT 5 has a context window of 400,000 tokens ~ 300,000 words.

This means if you’re using GPT 5 and the conversation goes beyond 300,000 words, the AI may “forget” or ignore the beginning.

If you’ve ever had an AI suddenly ignore something you said earlier in the conversation or start repeating itself, you might have hit the context window limit.

The model isn’t being dumb on purpose. It literally can’t see the text that scrolled out of its window. The context window includes both your input and the AI’s output, so a long-running chat can fill it up quickly.

Practical tip: To work around this, keep your prompts focused and to-the-point. If you have a big task, ask the AI to first break it down into logical sub-tasks, and then tackle each subtask as a separate prompt.

Knowledge Cutoffs - When AI's world stopped

The game has been going on inside a conference room for hours.

The host asks the next question: “Is it raining outside?”

The team has no idea. They can’t see outside. They don’t have access to weather apps.

But they can’t just say “we don't know”. The game requires them to give an answer.

So they do what they always do: they guess based on what they know.

Person 1 thinks about the season, the location, historical weather patterns, and writes “It”.

Person 2 continues with “might” and the chain continues until they produce:

“It’s probably raining outside based on typical weather patterns.”

Sounds reasonable. But it’s pure speculation because they don’t have the latest data.

Every AI has a ‘last updated’ date. If you ask about something that happened after that, it might guess, but it can’t know for sure.

GPT 4.1 has a knowledge cutoff date of Jun 01, 2024, and GPT 5 has a cutoff date of Sep 30, 2024.

To overcome this issue, the latest models have been given the capability of searching the internet if it feels that it may not have the accurate information for the answer.

Practical Tip: If you’re asking about very recent news, trends, or anything post-cutoff, always ask the model to check the internet so that it has the latest update.

Hallucinations - when AI confidently lies

Hallucination in AI doesn’t mean the computer is seeing pink elephants. It means the AI might generate an answer that sounds confident and plausible, but is completely fabricated or false.

Why do hallucinations happen?

Recall that the AI generates what seems like a probable answer based on patterns, not a database of verified facts. If the prompt is beyond its knowledge or ambiguous, it will still try to produce something – after all, its job is to respond.

This can lead to incorrect or made-up info. Think of it like a very eloquent person who hates to say “I don’t know,” so they always give you an answer, but sometimes they’re just winging it.

Hallucinations have reduced significantly since the intriduction of the web-search functionality. Still, always fact-check important details from reliable sources, especially if it’s anything data-driven or historical.

You can also prompt the AI in ways to reduce hallucinations, like: “Provide the source of this information”.

Now that you understand how AI actually works — predicting words one at a time with limited memory and knowledge cutoffs — you can work with these realities instead of fighting them.

Adopt the Right Mindset – AI as Your Partner, Not a Search Engine

Many beginners treat ChatGPT like a Google search box. They type a question and cross their fingers hoping for the best.

Treat the AI like an active collaborator rather than a one-shot answer machine. You bring the vision and context, it brings speed and structure in generating ideas and text.

It won’t replace your judgment or creativity, but it can dramatically extend your capabilities when you work with it. So approach it like you’re having a dialogue with a colleague: explain your needs, answer its questions, and refine the results together.

Remember, ChatGPT doesn’t “know” your world until you share information. So don’t hesitate to feed it background and specifics.

Treating the AI as a partner means teaching it about your business and goals, then letting it help you brainstorm and execute tasks.

Prompting 101: How to Write Effective Prompts

To ensure that your prompt contains all the elements the AI would need to produce a decent answer, make sure follows the R-T-C-G-O framework:

→ Role (R): Define who the AI should “be” for this task. Telling ChatGPT to “Act as an expert content marketer,” for example, can improve the relevance and tone of the answer.

Give it a persona or expertise to adopt. You can even mention the person you’d like it to behave as: “You are Dan Koe and you’re helping me establish myself as an authority in the personal development niche.”

→ Task (T) – What do you want to achieve? State clearly the outcome you’re looking for or the question you need answered. “Help me come up with an effective content strategy for the next 3 months”.

→ Context (C) - Give any background details the AI might need to answer better. Briefly explain your situation: your business type, audience, limitations (budget, time, skill level), or any info that’s relevant.

→ Guidelines (G) - Any specific guideline you’d like the AI to follow. “I want to focus just on my newsletter. No X or LinkedIn.”

→ Output Format (O) - The format you’d like the answer to be. “Format the response a as a numbered list, and keep each suggestion under 50 words.”

Bonus Tip: Use this master prompt to craft your prompts. This would ask you the required details and come up with a prompt following the prompting best practices.

Time-Saving Tricks and Prompt “Cheat Codes”

As you get comfortable, there are a few handy tricks to make the AI even more useful for you:

Use “ELI5” for simpler explanations

ELI5 stands for “Explain Like I’m 5.”

If you ask ChatGPT to “ELI5: [some topic],” it will break down the answer into very simple terms. This is great for understanding complex concepts without the jargon.

Ask for Step-by-Step instructions

Prepend “Step-by-step:” to your request or explicitly say “Give me the steps to accomplish X.” The AI will format the answer as a sequence of steps.

This is awesome for turning a daunting task into a manageable checklist. For example, “Step-by-step: How do I set up a basic Facebook ad?” would yield a numbered list of actions to take.

Use “TL;DR” to get summaries

If you have a long chunk of text (say a research article or a lengthy email), you can paste it and ask for a “TL;DR” summary. The AI will produce a concise synopsis. This is a quick way to digest long content.

Decision support with a “decision tree”

Typing “Decision tree:” before a question will prompt ChatGPT to lay out options and consequences.

For instance, “Decision tree: Should I hire a virtual assistant or continue doing tasks myself?” would result in a breakdown of pros and cons for each choice.

Build Your AI Toolbox & Workflow

To really get efficient, consider creating a little “AI playbook” for yourself. This can be as simple as a Google Doc or Notion page where you keep examples of prompts that worked well.

Over time you might develop prompt templates for common tasks. For instance, you might have a go-to prompt for blog outlines, another for social media captions, another for brainstorming product names. Save those!

Having a few ready-made templates means you don’t have to start from scratch every time.

Keeping Up Without Getting Overwhelmed

The AI field moves fast.

New tools, model upgrades, plugins, and features roll out seemingly every other week.

It’s easy to feel overwhelmed by the pace of change, but the good news is you don’t have to catch every new wave.

The fundamentals you learn today (like how to write a clear prompt and collaborate with the AI) will continue to serve you well, even as the tech evolves.

Focus on mastering the basics of communicating your needs to AI – that skill will transfer to any new AI assistant or feature that comes along.

And remember, you’re not alone if you ever feel frustrated or confused by AI.

Many entrepreneurs have felt the same at some point. The key is to approach it with a spirit of experimentation and learning. Over time, you’ll build confidence as you see AI become a reliable helper in your business.

Real-World Applications: How Can AI Help Your Solo Business?

Let’s end with some practical inspiration.

What kinds of tasks can a solopreneur actually offload or improve with AI? The answer is “almost anything involving text,” if you use it creatively.

Here are a few high-impact uses for ChatGPT and similar tools in a one-person business:

Content Creation and Copywriting

Whether it’s drafting blog posts, writing product descriptions, or crafting marketing emails, ChatGPT can save you hours.

You can have it generate a first draft, brainstorm catchy headlines, or rewrite a paragraph to be more engaging.

If you can think it up, ChatGPT can write it – from a witty social media caption to a rough draft of a business plan. Just remember to review and edit the output to add your unique voice before publishing.

Brainstorming and Idea Generation

Stuck on a name for your product or a topic for your next webinar? Ask the AI. It can generate dozens of ideas in seconds. Not every idea will be golden (some will be downright average), but they can spark your creativity.

For example, prompt: “Give me 10 creative podcast episode ideas about [your industry] targeting [your audience].”

You might only like 2 of them – and that’s still a win!

Research and Summarizing

Need to learn about a new market trend or digest a long report? You can ask the AI to explain a concept in simple terms, or to summarize a document for you.

Learning and Skill-Building

Think of ChatGPT as a mentor or tutor available 24/7.

You can ask it to explain how to do something step-by-step (e.g. “How do I set up a basic bookkeeping spreadsheet?”) or even to quiz you on knowledge you’re trying to learn.

If you don’t understand an answer, ask it to clarify. It’s patient and will rephrase until it clicks. This can boost your confidence in areas where you’re not an expert, by giving you a starting point to work from.

And that’s just scratching the surface. The core idea is to identify the tasks that typically suck up your time or energy.

Offload the small stuff so you can focus on high-level work.

Confidence Through Practice

AI might seem a bit overwhelming.

Start with the basics: approach ChatGPT with a clear goal, give it some guidance and context, and don’t be afraid to ask for exactly what you need.

If the result isn’t right the first time, that’s normal! Just refine your prompt or provide feedback and try again. Each time you do, you’re training yourself to communicate better with the AI (and often clarifying your own thinking in the process).

Over time, you’ll develop an intuition for how to phrase things and which details to include. Your prompts will get sharper, and so will the responses.

Dive in and experiment. With practice, ChatGPT (and other text AI tools) will become a natural extension of your solo business, helping you punch well above your weight class in terms of productivity and creativity.

A mediocre opening is a death sentence.

No matter how brilliant your newsletter is, if your first lines don’t grab readers by the throat, they won’t read another word.

10 seconds is all you have before they scroll away. Blow that, and all your research, expertise, and strategy vanishes into the void.

That’s why today I’m giving you a battle-tested template for writing openings to make readers stop and say: “This is something I can’t ignore.”

Subscribe now

The 5-Step Hook Formula

1. Lead with a strong stand

The first line must divide the room.

Take a position your reader can immediately agree or disagree with.

The inbox is a war zone. Every morning, your reader faces a flood of newsletters, promotions, and notifications.

If your opening line doesn’t spark emotion, curiosity, or tribal belonging, you’ve already lost.

Use confident, declarative language. Don’t hide behind words like “probably” or “maybe”.

Keep it short. Keep it punchy. Ten words or less.

Template:

[Target action/situation] is [pain word: exhausting / broken / overwhelming / a waste / painful / frustrating].

Examples:

• “Crafting prompts is a pain.”

• “Building authority online is exhausting.”

• “Creating content consistently is a grind.”

2. Describe the broken approach your reader takes and the struggles they face

Now, show them the mistake they’re making. Mirror their behavior so precisely they feel caught in the act.

The purpose of this sentence is to make the reader nod and think “Yes, that's exactly what I do!”

It should contain two important elements:

• the current action they take

• which results in the specific struggle they experience

Template:

“You [specific current action], trying to [what they're attempting] and [what they need].

Most of the time you end up [suboptimal result] or [time-wasting behavior] till you [eventual outcome].”

Example: “You stare at that blank input box, trying to figure out where to begin and what to type to get the results you need.

Most of the times you end up using generic prompts that produce mediocre content or wasting hours fighting with the AI, going back and forth till you find something usable.”

3. Highlight the consequences of following this approach

Show the cost of staying stuck.

Template:

“[Waste/Loss/Negative outcome] and [additional consequence].”

Examples:

• “Lots of wasted time and effort.”

• “Burnout without the breakthrough.”

• “Hours lost and nothing to show for it.”

• “Endless back and forth with mediocre results.”

• “You publish yet another newsletter that gets ignored.”

4. Paint the picture of a life without this problem

Create hope and desire for the solution

Template:

“What if you could [desired outcome] without [current struggle]?”

Examples:

• “What if you could create authority content in 90 minutes instead of 5 hours?”

• “What if you could turn your expertise into consistent content without losing your voice?”

• “What if you could generate a baseline prompt for any writing task without facing the blank page problem?”

Important: Never overpromise. Be plainspoken and value clarity above hyperbole.

5. Promise the specific tool/system that makes the vision real

Close the loop and promise the tool or system that makes the vision real.

Template:

“Today, I’m sharing [The System Name] — [what it does] and [specific benefit].”

Examples:

• “Today, I’m sharing The Master Prompt — a prompt that acts as your personal prompt engineer and creates structured prompts for your specific needs.”

• “Today, I'm sharing The Content Assembly Line — a system that transforms your expertise into published content in 90 minutes flat.”

• “Today, I’m sharing The Authority Engine — a workflow that turns your scattered knowledge into strategic content that builds your reputation.”

The Hook Generator Prompt

This prompt crafts a compelling opening for any topic.

When you use it, the AI will ask you a few questions about the topic, and then craft an opening you can use as your starting point.

Instead of starting from scratch, you’ll get a solid baseline to build on.

<Role>: You are a world-class copywriter who specializes in writing newsletter openings that hook readers instantly. You think in terms of persuasion, psychology, and storytelling, not fluff. Your job is to write compelling introductions that make the reader nod, feel the pain, crave the solution, and keep reading.

<Task>: I will give you a topic, audience, and main promise. Write a high-converting newsletter opening using the 5-Step Hook Formula provided below. Keep the copy sharp, conversational, and emotionally charged.

<Guidelines>:
Before you create the opening, ask for things you would need to make sure you do a good job, like the main topic/subject of the newsletter, the main pain the reader faces that the article addresses, and the promise made to the reader. Ask follow-up questions if you need clarity, but don't ask too many.

<5-Step Hook Formula>:
1. Strong Stand: "[Target action/situation] is [strong emotional word like: exhausting/broken/overwhelming/painful/frustrating]."
This is the opening line. Take a position the reader can immediately agree or disagree with. Use confident, declarative language. No hedge words like “probably” or “maybe”. Keep it short and easy to read (10 words or less).

Examples:
- "Crafting prompts is a pain."
- "Building authority online is exhausting."
- "Creating content consistently is a grind."

2. Current Struggle: "You [specific current action], trying to [what they're attempting] and [what they need]. You end up [suboptimal result] or [time-wasting behavior] till you [eventual outcome]."
Mirror the behaviour of the reader so closely that they feel caught in the act. The purpose of this sentence is to make the reader nod and think “Yes, that's exactly what I do!”

It should contain two important elements:
- the current action they take
- which results in the specific struggle they experience

Example:
"You stare at that blank input box, trying to figure out where to begin and what to type to get the results you need.
You end up using generic prompts that produce mediocre content or wasting hours fighting with the AI, going back and forth till you find something usable."

3. Costly Result: "[Waste/Loss/Negative outcome] and [additional consequence]."
This sentence shows what it costs them in time, energy, or growth and creates an urgency for change.

Examples:
- “Lots of wasted time and effort.”
- “Burnout without the breakthrough.”
- “Hours lost and nothing to show for it.”
- “Endless back and forth with mediocre results.”
- “You publish yet another newsletter that gets ignored.”

4. What If Vision: "What if you could [desired outcome] without [current struggle]?"
This sentence paints a vision for the future without this problem. This also sets up the promise the article makes to the reader.
Important: Never overpromise. Be plainspoken and value clarity above hyperbole.

Examples:
- “What if you could create authority content in 90 minutes instead of 5 hours?”
- “What if you could turn your expertise into consistent content without losing your voice?”
- “What if you could generate a baseline prompt for any writing task without facing the blank page problem?”

5. Solution Promise: "Today, I'm sharing [The System Name] — [what it does] and [specific benefit]."

Examples:
- "Today, I'm sharing The Master Prompt — a prompt that acts as your personal prompt engineer and creates structured prompts for your specific needs."
- "Today, I'm sharing The Content Assembly Line — a system that transforms your expertise into published content in 90 minutes flat."
- "Today, I'm sharing The Authority Engine — a workflow that turns your scattered knowledge into strategic content that builds your reputation."

<Writing Rules>:
- Write in short paragraphs (1–2 sentences max).
- Use visceral, plainspoken language. No jargon.
- Sound confident and direct. Never hedge.
- If there are multiple possible framings, provide 2 variations so I can choose.
- Optimize for mobile: no big blocks of text.

<Output Format>:
- Write the final opening as if it’s ready to drop into a newsletter.
- One sentence paragraph is okay. Never go beyond 2 sentence paragraphs.

How to Use the Hook Generator (With Real Examples)

Let me show you exactly how to use this prompt with a few examples.

Example: Crafting an opening for a post on workout consistency

Let’s say you’re teaching a framework that helps busy professionals stay consistent with workouts.

Step 1: Paste the Hook Generator into ChatGPT/Claude

The AI will ask a few quick questions about your post, like:

Step 2: Provide the required details

The AI will then generate an opening using the 5-step framework:

Step 3: Work with the AI till you get something you like

The Hook Generator gives you a solid, structured opening that hooks the reader following the 5-step formula — so you never start from a blank page again.

Now it’s your turn. Fire up the Hook Generator, feed it your topic and audience, and see what it gives you.

Creating content consistently can be a real grind.

Research → outline → write → edit → format → publish → repeat.

While most of us burn out chasing inspiration and grinding through manual processes, a few thrive, writing online and building authority consistently for years.

The difference? Systems.

Take Justin Welsh for example, one of the most authoritative voices on solopreneurship. He’s been writing online almost daily for the last 6 years!

Here’s what he has to say:

Build systems that make success inevitable

It’s Thursday night and you still have no idea what to write about for Friday’s newsletter. You frantically scroll through articles and research reports, hoping something will spark an idea.

Next week, same panic. Same last-minute brainstorming session.

Every single time you’re reinventing the wheel from scratch.

Lots of wasted time and effort.

You’re already consuming the right information — industry reports, resesarch papers, case studies. The raw material for creating thought leadership content is right there.

What you lack is a system — a structured way to generate ideas.

What if you had a system that automatically generated ideas from the information you consume?

Today, I’m sharing the Article-to-Ideas System — a simple workflow (which you can download for free) that generates content ideas tailored to your brand automatically.

Subscribe now

What is a workflow?

A workflow is simply a sequence of steps to complete a task.

Let’s say you’ve come across a fascinating industry report containing insights that could spark weeks of content ideas.

Here’s what you typically do:

1. Go to ChatGPT

2. Upload the PDF

3. Describe your brand and audience

4. Ask for 10 newsletter ideas based on the report

5. Wait for AI to generate the ideas

You just executed a workflow — manually.

If you think closely, your presence and decision-making was nowhere required in the whole process.

1. Go to ChatGPT → The link to ChatGPT remains same everytime. No decision-making required.

2. Upload the PDF → The process to upload a document to ChatGPT remains same everytime. No decision-making required.

3. Describe your brand and audience → Your brand details stay the same. No decision-making required.

4. Ask for 10 newsletter ideas → The prompt remains same each time: “generate 10 newsletter ideas based on the uploaded PDF”.

5. The AI generates article ideas → The AI does the job. All you do is sit back and relax.

Here’s the surprising part:

You don’t add value by doing this process manually. You just become the bottleneck.

The system already knows what to do. It just needs to be run.

And if that’s the case — why not have it run automatically?

Let me show you what that looks like.

Overview of the Article-to-Ideas System

We’ll use an automation platform called n8n (pronounced “n-eight-n”) to automate our content ideation task.

n8n let us connect different modules together, each performing one subtask in the workflow.

1. You upload the PDF of an article/industry report, and mention the details of your brand (your niche and target audience)

2. Step 2 extracts all the text from the PDF

3. Step 3 creates a custom prompt based on your brand and contents of the PDF

4. Step 4 sends the PDF contents and brand details to the AI to generate article ideas

5. Step 5 formats the AI’s response

6. Step 6 displays article ideas

I’ve intentionally kept this workflow very basic for ease of understanding. We can easily extend this to run automatically when you drop an article PDF in your Google Drive, and have it generate article ideas using AI and store them in a Google Sheet.

You can then create another workflow that takes an idea from the Google Sheet and drafts an article for you.

This is how you create automated systems that make your life easier, and success inevitable.

If you perform any task where no cognitive input is required, break it into small steps, and build a system that automates the entire process.

Note: n8n is a paid platform but comes with a nice 14 day trial. There are other automation platforms as well like Make and Zapier, but I find n8n to be the best.

Okay, workflows sound nice. But do I need a technical background to implement this?

Not at all.

I know these automation tools can seem intimidating at first, but this workflow requires zero coding or technical expertise.

If you get stuck anywhere, just upload the screenshot to ChatGPT and it will fix it for you.

To implement this workflow, you’ll be doing five simple things:

1. Download the workflow using the download link

2. Create a free n8n account (takes 2 minutes)

3. Import the downloaded workflow file

4. Create your OpenAI API key (takes 5-10 minutes)

5. Add your OpenAI API key to the workflow (takes 2 minutes)

The whole setup takes less than 15 minutes.

Setting up your Article-to-ideas workflow

Step 1: Download the workflow

Click here to download the workflow. It will download a file called 001_Article_to_Ideas.json

Step 2: Create your free n8n account

1. Go to n8n.io

2. Click “Get Started” in the top right corner

   

3. Fill up the details and click on “Start 14 day trial”

   

4. Fill up the questionnaire

5. Skip invite

   

6. Click on “Start Automating”

   

7. In the dashboard, click “Create Workflow”

   

8. You’ll be taken to the workflow editor. You’re now ready to import the Article-to-Ideas System.

Step 3: Import the Workflow

Click on the three dots at the top of the workflow editor → select “Import from File…” → Select the file you just downloaded

This will create the workflow in your editor:

Step 4: Create your OpenAI API key

An Application Programming Interface (API) acts as a bridge that enables different software programs to communicate with each other.

Normally, when we interact with ChatGPT, we manually type in the commands and read the responses.

This is okay for one off conversations.

To be able to connect with OpenAI’s GPT models automatically, you’ll need an API key.

Important Note: API usage is billed separately from any ChatGPT Plus subscription you might have.

You’ll need to add a payment method to your OpenAI account, and you’ll be charged based on your API usage.

For the system we’re building, costs are typically minimal (~$5 per month for moderate usage).

Follow the steps here to generate an API key.

Copy the API key. We’ll need it for the next step.

Step 5: Add the API key to your workflow

Come back to n8n and add the key you just created.

1. Click on the “Generate ideas using OpenAI API” node

2. In the panel that opens, look for “Credential to connect with”

3. Click on “Create new credential”

4. Paste the API key you copied earlier

Alright! The workflow is ready and we’re ready to roll.

Test Your Article-to-Ideas System

Time to see your automation in action.

1. Click on “Upload PDF and describe your brand” → “Execute workflow”

   

2. A form will appear asking for the PDF file, your niche description, and your target audience details

   

3. Click “Generate Ideas”, and the workflow will process the PDF and generate a list of article ideas tailored specifically to your brand based on the document you uploaded.

Understanding the workflow

Beyond Article Ideas: Your Systems Journey Begins

Setting up the workflow might seem complicated, but you’ve just crossed a critical threshold.

You’re no longer someone who panics about content every Thursday night. You’re someone who builds systems that work while you sleep.

Next week, we’ll extend this workflow to automatically generate ideas and store them automatically in Google Sheets.

Here's what becomes possible when you think systematically

Research-to-Authority Pipeline:

• Save a McKinsey report to Google Drive → wake up to 10 article ideas in your inbox, each positioned to showcase your unique expertise

• Bookmark competitor analysis → automatically generate contrarian takes that establish you as an independent thinker

• Forward client success stories → transform into case study content that builds credibility across LinkedIn, newsletters, and blog posts

Automated Content Operations:

• Google Drive PDF uploads → automatic idea generation → organized content database

• Research compilation → multi-format content → cross-platform distribution

• Expert insights → systematic content → consistent authority building

Getting errors?

Copy any error message and ask ChatGPT: “I’m setting up an n8n workflow and got this error. How do I fix it?” It’s surprisingly effective at troubleshooting automation platforms.

AI feels overwhelming.

Every week brings new tools, tutorials, and “revolutionary” breakthroughs. You know AI could help your business, but you don’t know where to start.

Here’s what most solopreneurs get wrong: they think they need to be on top of the latest AI tool in the market to be on top of the AI game.

The truth is, you don’t need to learn a hundred different tools.

You just need to learn how to use ChatGPT or Claude effectively.

Today, I’ll show you how I use Claude to automate my processes using a real example, then help you adapt this to your specific business processes.

Subscribe now

Step 1: Think about what existing process should be improved

Instead of asking, “What AI tools should I use?”, ask: “What do I do repeatedly that takes too much time?”

Anything that you perform repeatedly and follows a logical sequence of operations is a good candidate to be automated using AI.

Let me show you exactly how this works.

I watch lots of podcasts to stay on top of the industry. I take detailed notes to understand topics and generate article ideas. These notes help me understand topics thoroughly and generate article ideas for my newsletter.

The watching is fun, but the note-taking is time-consuming and messy.

Sometimes I miss critical points, and the perfectionist in me worries about what other valuable insights I might have missed.

I’d love to have an assistant that would automate the note-taking for me.

While there might be tools out there that could do this, I want something that’s tailor-made for me so that I can control exactly how it behaves.

Step 2: Break a high-level task into atomic low level tasks

Don’t ask the AI: “Help me automate note-taking from YouTube videos.”

While the AI would try its best to fulfil the request, it’s highly unlikely that it would do a good job. Why?

This is a high-level task with countless nuances:

• What kind of YouTube video is it? Podcast, instructional, motivational?

• What’s the goal of note-taking? Deep understanding or idea generation?

• Where should notes be captured? Google Docs, Notion, Obsidian?

• What specific elements should every note contain?

For example, here’s what I think every podcast note should include:

• Quotes: Interesting quotes made by speakers

• Keywords: Key terms mentioned in the video

• Questions Asked: Questions posed during the conversation

• Concepts Explained: Core concepts that were broken down

• Article Ideas: Potential newsletter topics based on the content

• Mind-blowing Revelations: Moments that made me go “Whoa!”

• Other Noteworthy Points: Additional insights worth capturing

Without these details, AI will produce something generic and largely useless.

How to break a high-level task into low-level tasks

Here’s what I would do:

Open Claude/ChatGPT and type:

I run a newsletter called DeepWriting (DW) that helps solopreneurs looking to build authority online create AI-powered content systems that turn expertise into consistent, authority-building content in a fraction of the time.

I watch a lot of YouTube podcasts and take down notes. These notes are used to thoroughly understand the topic(s) and to come up with article ideas for my newsletter. Watching the videos is fun but taking down notes is time-consuming. Also, sometimes I miss an important point.

I want to use AI and automatically generate a thorough note from a YouTube podcast.

I want you to break down the whole process of going from a YT podcast to a structured note into logical atomic steps. Do not think about tools to implement a step now.

Also, do not assume anything. If you have any doubt, ask questions.

This step is meant to brainstorm with the AI and break down the high-level goal into smaller steps. This step often makes me think about a process differently.

Claude asked me clarifying questions like:

After answering, it broke down the high-level task into subtasks:

Step 3: Ask the AI to research and find tools for each step

Now that the steps are finalized, let’s ask the AI to list the tools that will help us automate the whole process. Here’s a sample prompt I use:

Perfect! Let's go step-by-step. Search the web and suggest a tool for Step 2: Transcript Extraction. This tool should be able to input the audio file and output transcripts with high accuracy. It should have a generous free tier and satisfy most of our requirements.

This is what Claude gave:

Iterate and ask questions till you understand why we’re using a tool and why not something else. Consider constraints like budget, scale, and ease of use.

Step 4: Once the tools are finalized, ask AI to guide you with implementation

After you’ve identified all the tools for each step, ask the AI to help you build the system. Start with simple connections between tools, then gradually add complexity.

The key here is to build incrementally. Get one step working before moving to the next.

This is what I built after working with the AI in one evening:

End-to-end workflows in n8n that:

1. Takes a YouTube video mp3 as input from Google Drive (I extracted the mp3 out of the system for now)

2. Transcribes the audio using Deepgram API

3. Stores the transcription into a specific folder in Google Drive

4. Picks up the transcription file from the Google Drive

5. Extracts my brand details like value statement, ideal customer persona etc. from a database

6. Generates article ideas using Claude and ChatGPT and stores them back in the database

7. Score each article idea indicating how likely they are to resonate with my audience

My knowledge of coding did help me here, but even without coding, one can implement this in 2 days, working with the AI.

Remember: “Don’t hesitate to ask stupid questions”. A few months back I had no idea what n8n was, and I just asked questions like: “What is n8n?”, “How difficult is it to learn?”, “How do I connect to my Gmail using n8n?”

Step 5: Stay in the loop and iterate

Run your automation on 3-5 pieces of content first. Compare the AI output with what you’d create manually.

Ask yourself:

• Is the AI missing important details I’d normally catch?

• Is it capturing information I don’t actually need?

• Are the outputs in the format that’s most useful for your next step?

Make incremental improvements

Don’t try to fix everything at once. Pick the biggest problem and solve that first.

For my podcast note-taking system, I noticed the AI was capturing too many minor details and missing the “aha moments” that matter most for article ideas.

I solved this by adding a scoring step that rates each insight based on how relevant it is to solopreneurs building authority.

Ask AI to suggest optimizations

After a few iterations, copy a few examples of your automation’s output and ask the AI:

Here are 3 examples of notes my automation created. What patterns do you see? What could be improved to make these more useful for generating newsletter article ideas?

The AI often spots patterns you’ll miss.

Scale gradually

Start with your most frequent, time-consuming process. Once that’s working smoothly, identify your next automation candidate.

Don’t try to automate everything at once. Master one system, then build the next.

How to adapt this to your business

The exact same approach works for any repetitive process in your business.

For client onboarding: Break down your current process (initial call → contract → project kickoff → deliverables). Ask AI to suggest tools for each step. Build connections between them.

For content repurposing: Map out how you currently turn one piece of content into multiple formats. Identify the manual steps. Find AI tools that can handle the transformation and formatting.

For lead follow-up: Document your current sequence (initial outreach → follow-up emails → proposal → closing). Use AI to draft personalized messages and automate the scheduling.

The framework stays the same:

1. Identify the repetitive process

2. Break it into atomic steps

3. Research tools for each step

4. Build incrementally

5. Monitor and improve

Bottom line

You don’t need to become an AI expert overnight.

You don’t need to learn dozens of new tools.

Start with ChatGPT or Claude. Pick one time-consuming process. Break it down. Build it step by step.

In a few weeks, you’ll have your first automated system running. Then you can tackle the next one.

The solopreneurs winning with AI aren’t the ones using the most tools. They’re the ones who've automated their most painful processes.

The old way of selling online is dead.

Bombarding people with ads and hoping something sticks is an expensive game.

The new path to revenue isn’t shouting louder than your competition. It’s establishing genuine authority in your niche.

And the simplest way to establish authority online is to become the informational source for your ideal customer.

When you create a body of content that proactively solves the pain points your ideal customer has, or teaches them something that makes their lives better, they feel connected and are much more likely to respond to your offer.

This is exactly how HubSpot stood out in a highly saturated and competitive SaaS market — by creating high-quality, in-depth content.

And once you establish yourself as the go-to authority in your niche, everything in your business becomes easier.

• You no longer chase clients. They seek you out.

• Your sales conversations shift from convincing people of your value to discussing how you’ll work together.

• You can charge premium prices because the ideal customer already recognizes your expertise.

Every entrepreneur worth their salt knows this.

They start creating content with lots of enthusiasm, with the goal of becoming an authority in their niche.

Yet most of them never get there. Why?

Subscribe now

99% of online creators are reactive

They chase trends, react to random ideas, and write about whatever inspires them that week.

I call them content hamsters, forever running on the hamster wheel of reactive creation, starting from zero each time thinking “What should I write about this week?”

As a result, their content feels scattered, their audience feels confused, and they can’t point to a clear territory they own.

Six months on, they’re frustrated. The growth plateaus despite working twice as hard, until they eventually burn out and quit.

The fact is, they make content creation a lot harder than it needs to be.

Only 1% of creators online create with intention and strategy

I call them authority builders.

What I’ve learned studying successful creators is that authority isn’t just about creating good content.

It’s about systematically covering a territory.

Hang on a sec, it’ll become clear.

Authority builders understand that each piece of content serves a specific purpose.

They might publish less frequently, but everything they create strategically builds their position as the go-to expert.

Their content forms a cohesive body of work that naturally leads prospects toward their offers.

They work from systems, not inspiration.

Marking your territory, or as the geeks call it: “identifying your topic taxonomy“

Think about someone you consider an authority in their field.

Go through their content carefully, and you’d see they organize their content into 4-5 broad themes.

These are the things they wish to be known for, also known as content pillars.

Take Justin Welsh, one of the most authoritative voices on solopreneurship as an example.

His content is strategically organized around six pillars:

1. Personal Growth

2. Business Planning

3. Personal Branding

4. Social Media Growth

5. Monetizing Knowledge

6. Productivity & Time Management

Similarly, James Clear divides his content into 8 authority pillars:

1. Creativity

2. Decision Making

3. Focus

4. Habits

5. Life Lessons

6. Motivation

7. Productivity

8. Self Improvement

Once these top creators identify the major themes, they define 5-10 specific topics under each pillar that their ideal customers genuinely care about.

These are more specific, and often address particular pain points or aspirations.

For example, Justin’s “Business Planning” can be broadly classified into these sub-pillars:

1. Finding a profitable niche

2. Crafting a business roadmap

3. Becoming a solopreneur

4. Solopreneur business models

5. Defining your mission and goals

This carefully mapped territory (topic taxonomy) covers everything his ideal audience needs to know about building a successful one-person business.

Each piece of content he publishes strengthens his claim on this territory.

When someone discovers his work through a LinkedIn post about “finding your profitable niche,” they naturally discover his other content about business planning, social media growth, and monetization.

The result?

Justin is now recognized as one of the definitive authorities on building a solopreneur business online, making millions in digital product sales, all without spending a dime on paid advertising.

Now that you know what topic taxonomy is, and why it’s important, let’s see how to create yours using AI.

How to create topic taxonomy using AI

This prompt works best with Deep Research models like ChatGPT’s Deep Research and Gemini Deep Research.

Open up a new research window and paste the below prompt:

<Role>:
You are an expert market researcher tasked with generating a carefully structured topic taxonomy specifically tailored to my brand. The goal is to establish myself as the go-to authority in my niche by creating strategic, connected content that positions me as the definitive resource for my ideal customers.

I will provide the following detailed brand documents for your analysis:

1. Unique Value Proposition (UVP)
2. Ideal Reader Persona (audience demographics, behaviors, goals, interests)
3. Key Pain Points faced by the Ideal Reader

<Task>:
Your goal is to deeply research, analyze, and propose a comprehensive topic taxonomy to guide content ideation and ensure alignment with the brand’s UVP and the ideal audience’s needs and pain points.

Structure your taxonomy clearly as follows:

Level 1 (Pillar): Identify 4-6 primary pillars that directly align with the brand’s UVP and strongly address the audience’s pain points.

These should be broad enough to encompass multiple sub-topics but specific enough to claim a distinct territory.

Provide a brief rationale explicitly connecting each pillar to the brand’s UVP and the ideal reader's identified pain points.

Level 2 (Sub-pillars/Clusters): Under each pillar, propose 5-10 focused sub-pillars or clusters.

These should address specific pain points, questions, or aspirations my audience has.

For each sub-pillar, include a concise rationale that clearly links it to both its parent pillar and audience pain points.

Level 3 (Specific Topics/Questions): Within each sub-pillar, list 3-5 specific, actionable, and engaging topics or questions that resonate with the ideal audience and address their specific concerns or interests.

<Important Guidelines>:
* Your output must be based entirely on the provided brand documents.
* Clearly state your rationale for each level, linking explicitly back to the UVP and audience pain points.
* If the provided documentation is unclear or insufficient for confident taxonomy generation, explicitly state any uncertainties and request additional clarifications from the user rather than guessing.

<Output Format>:
Present your findings strictly in the following structured format:

Level 1 – Pillar #1: [Name of Pillar]
- Rationale (briefly connect pillar to UVP & pain points)
  - Level 2 – Sub-pillar #1:
    - Rationale (connection to Pillar & audience pain points)
      - Level 3: Specific topics/questions:
        1. [Specific topic/question]
        2. [Specific topic/question]
        3. [Specific topic/question]
  - Level 2 – Sub-pillar #2:
    - Rationale
      - Level 3: Specific topics/questions:
        1. ...
        2. ...
        3. ...
(Continue clearly through all pillars)

<UVP>:
[Paste your UVP here]

<Ideal Reader Persona>:
[Paste your ideal reader persona here]

<Key Pain Points>:
[Paste your ideal reader's pain points here]
#####
Please begin by thoroughly analyzing the provided documents and request clarification immediately if you encounter insufficient detail or ambiguity to ensure maximum accuracy and relevance of your taxonomy proposal.

Paste your UVP, Ideal reader persona, and their pain points at the designated places.

Check this article in case you need help creating any of the above.

Answer any follow-up questions the model asks and wait for the results.

Here’s a pillar it generated for a hypothetical business that sells turmeric-based health drink:

The pillars are broad and evergreen, and each has a few specific sub-themes.

For instance, under Lifestyle & Recipes, sub-topics like “pre-workout mocktails” or “golden-milk lattes” indicate recurring content ideas (perhaps a series of recipe posts or lifestyle tips centred on those themes).

Notice how each sub-theme is narrow enough to be interesting, but not so narrow that you’d run out of content ideas.

Pretty solid!

How to use your territory map to build authority

Store your taxonomy in a document and use it in your next ideation session.

The question is no longer “What should I write about this week?”

You simply look at the map and ask “Which part of my territory needs strengthening right now?”

Do not just ask the AI to produce article ideas. Specify the pillar and subpillar you’d want to create content around.

The aim is to produce enough content under each pillar that someone exploring your site could get a full education on that theme.

When creating content, use your taxonomy to create meta tags and on-page SEO. Include pillar keywords in category pages, and sub-topic keywords in titles and headings.

Iterate

Remember, this taxonomy is just the first draft. Over time, you may need to tweak your taxonomy. Maybe one pillar isn’t performing or you discover a new pillar emerging from audience feedback.

Schedule periodic check-ins to evaluate if your taxonomy still reflects your content and audience interests. If you pivot your business, the taxonomy should pivot with you.

From scattered content to genuine authority

Here’s what happens when you implement a strategic topic taxonomy:

1. Your content creation becomes effortless. No more wondering what to write about. Your map guides you.

2. Your audience starts recognizing patterns in your content. They begin to associate you with specific expertise territories and return to you for more information.

3. Search engines start trusting your site on certain topics. When a website covers a subject in great detail through many interrelated pieces, Google starts viewing it as an authority on that subject.

You transform from just another content creator into a recognized authority, someone whose opinion matters, whose expertise is valued, and whose offers command premium prices.

So ask yourself: Are you still running on the content hamster wheel, or are you ready to start building with intention?

It’s you against a prompt box.

You type in your request, and hit send. What comes back is... disappointing.

I’ve seen even experienced AI users make these 5 mistakes that kill productivity and waste time.

In this article, I’ll provide straightforward fixes you can apply immediately.

Let’s begin!

Subscribe now

1. Not specifying the target audience and their knowledge level

When your prompt lacks specificity, the AI makes random assumptions about your audience, their knowledge level, and what information would be valuable to them.

Answer these two questions:

1. Who are we writing for?

2. What do they know?

Bad prompt

Write a blog post about automating content systems using AI.

Better prompt

Write a 1,200-word blog post about automating content systems for solopreneurs who want to scale their business using AI.

Assume they understand the basics of AI but don't know how to use it to automate their workflow.

2. Not specifying what the goal of the article is

To build authority online, every piece of writing must address a pain point the ideal reader has, and make their life tangibly better.

1. What’s the pain point we’re solving?

2. What’s the promise we’re making to the reader?

Bad prompt

Write an article about coming up with content ideas using AI.

Better prompt

Write a 1200-word article about coming up with content ideas using AI.

Here are the required details:

1. Target audience: Solopreneurs looking to build authority online through written content.
2. Knowledge level: They understand the basics of AI but don't know how to use it to automate their workflow.
3. Pain point resolved: Every week, they're faced with the same dilemma: what should we write about?
4. Promise made: The article should give them a practical system to automatically come up with content ideas using AI. The reader should be able to implement this system within a week and using free tools.

3. Using the same model for different tasks

Different models excel at different tasks.

• Reasoning models (like ChatGPT’s o3/Claude Extended Thinking) help you think deeply about a problem and break it down to smaller sub-problems. Use these models when you have a rough idea of an article and need help brainstorming.

• Research models (like ChatGPT’s DeepResearch) excel at researching a topic. Use these when you have an idea of the topic you want to write about, and want to conduct an in-depth research.

• Regular models (like Claude/4o) perform better at normal writing tasks. Once you know the topic you want to write about, and have the related research, provide them to a regular model and ask it to write for you.

4. Jamming too many tasks in one prompt

Cramming multiple content elements into a single prompt creates the “mile-wide, inch-deep” problem.

Don’t throw the kitchen sink at the AI. Break big tasks into smaller sub-tasks.

Let’s say you want to create an article on the common prompting mistakes people make while creating content with AI.

Don’t ask the AI to

Write a 1200-word article about the common prompting mistakes people make while creating content using AI.

Instead

Break the task down to individual components, and use separate chat sessions:

Task #1: Research the common mistakes people use while using AI to create content

I need to write an article about common prompting mistakes people make when creating content with AI.

Research and identify the 5-7 most critical mistakes that impact content quality. Understand the industry best practices and the latest research. For each mistake, explain its negative impact on the resulting content and how to fix it.

Task #2: Write the article tailored to your audience. Include the research in the earlier step as part of the prompt.

Write a 1200-word article about the common prompting mistakes people make when creating content with AI.

Here are the required details:

1. Target audience: Solopreneurs looking to build authority online through written content.
2. Knowledge level: They understand the basics of AI but are not aware of the prompting best-practices.
3. Pain point resolved: New users often struggle with getting useful answers from AI because they don’t know how to ask effectively.
4. Promise made: Instead of feeling that an AI reply is a hit-or-miss surprise, they’ll feel more in control and capable of steering the conversation.
5. Research: Here's the research conducted on the topic

Task #3: Edit the article to sound like you.

Here's an article draft on the common prompting mistakes people make when creating content with AI.

Edit this article draft to match my unique voice. Ensure that it follows the style guide to the point.

Here's the style guide:

[Insert your style guide]

Here’s how you can create your brand’s style guide.

Perform each of these tasks in separate chat sessions, and choose the appropriate model for each.

5. Starting from a blank screen

It drains energy to craft a prompt from the scratch every single time. Instead, start with a pre-made prompt and tweak as per the requirements.

• Use this master prompt to generate the starting prompt

• Give it the information we discussed in points #1 and #2

• Check if the prompt contains all the necessary details. Tweak as required.

Targeting the wrong customers kills businesses.

You’ve created something valuable or have an idea for a great product or service. You’re ready to launch.

But now comes the million-dollar question: Who exactly should you target, and what should you say to them?

This uncertainty leads to

• Wasted time creating content for audiences who won’t convert

• Developing offers that don’t resonate

• Pricing services without market context

• Building credibility with people who’ll never purchase

The solution? Strategic market research powered by AI.

In this guide, I’ll show you exactly how to use AI to identify your highest-ROI target personas, analyze their pain points, and develop messaging that resonates.

By the end, you’ll have a data-backed strategy for launching your product with confidence, even with limited resources.

Subscribe now

The AI-Powered Market Research Framework

This framework provides clear direction on targeting and messaging for businesses of any size, but it’s especially powerful for small businesses with budget constraints.

To conduct our market research, we’ll use the Deep Research feature in tools like ChatGPT/Gemini/Perplexity.

These tools take a research task, break it down into sub-tasks, search the internet for required information, and compile them into a report.

We will specifically ask the tool to analyze:

1. The market trajectory - Is the market growing, stable, or shrinking?

2. What the top performers in the domain are doing, their pricing strategies, and differentiation

3. Analyze 3-5 personas the product can target and come up with the highest ROI persona, given our resources and constraints

4. The optimal pricing strategy to adopt

5. The pain points your target persona faces and messaging that will resonate

6. The most effective market entry strategy

Say you’re a solopreneur looking to launch a consulting business.

• You work alone and are short on capital.

• You want to target the highest ROI persona through content, with the right messaging.

• You need to know the exact steps to generate cash quickly, while working on a clear path to scale in the long term.

Let’s walk through the exact process of using AI to create your market research strategy.

Step 1: Generate The Deep Research Prompt

To get the best results out of Deep Research models, we need to tell them exactly what we need.

That’s why I created a meta-prompt that generates the perfect Deep Research prompts for you.

Important: Use this prompt with a reasoning model like ChatGPT 3o/Claude Extended Thinking for best results.

<System>
You are an expert prompt engineer. Your task is to craft a single, high‑fidelity Deep‑Research prompt that will generate a McKinsey‑style market intelligence report enabling a business to:

1. Detect emerging market trends.
2. Identify and justify the highest‑ROI customer persona considering their constraints (team size, budget, expertise) and experience.
3. Craft pricing via Van Westendorp analysis.
4. Pinpoint persona pain points and resonant messaging channels.
5. Lay out an authority‑building content plan.

You do **not** do the research yourself; you only create the prompt. You may ask concise follow‑up questions when essential details are missing. Otherwise, minimise friction.
</System>

<Workflow>
1. Greet the user and request the following information (bullet list):  
   – Product / service description
   – Industry / niche
   – 3-5 target personas (short labels)
   – Geographic focus (if any)
   – Budget: Approximate launch budget and expected monthly marketing spend.
   – Team: Current head‑count and each member’s key skills (e.g., SEO, copywriting, design, dev, analytics, domain expertise).
   – Owned Marketing Channels & Audiences: Email list size, social followings, community/forum access, partnerships, etc.
   – Revenue Time Horizon: How long can they operate before seeing the first meaningful revenue (in weeks or months)?
   – Non‑Negotiable Limitations: Examples: “no paid ads,” “must stay within EU data‑privacy rules,” “only English‑speaking markets,” or any capacity/time constraints.
   – Key differentiators or assets
2. Wait for the user’s answers. If any item is missing or unclear, ask a single clarifying question.
3. Once all inputs are collected, inject them into the **Prompt Template** verbatim.
4. Return **only** the completed Deep‑Research prompt, wrapped inside triple back‑ticks ``` for easy copy‑paste.
5. End the interaction.
</Workflow>

<Constraints>
• Do not add commentary outside the final triple‑back‑tick block.  
• Preserve all headings, numbering, and wording in the template.  
• If concrete market data is requested but unavailable, instruct Deep Research to mark as “Data unavailable – recommend primary research”.  
• Keep any follow‑up questions as short as possible.
</Constraints>

<Prompt Template>
<System>
You are an elite market‑intelligence analyst producing McKinsey‑calibre reports with current data and explicit citations. When uncertain, you ask clarifying questions rather than assume.
</System>

<Context>
• Product / Service: {{PRODUCT_DESCRIPTION}}
• Industry / Niche: {{INDUSTRY}}
• Geography: {{GEOGRAPHY}}
• Candidate Personas:
  1. {{PERSONA_1}}
  2. {{PERSONA_2}}
  3. {{PERSONA_3}}
  4. {{PERSONA_4}}
  5. {{PERSONA_5}}
• Resources: {{RESOURCES}}
• Constraints: {{CONSTRAINTS}}
• Expertise: {{EXPERTISE}}
• Differentiators / Assets: {{DIFFERENTIATORS}}
</Context>

<Instructions>
1. **Market Trajectory**
   – Quantify TAM, SAM, SOM and 5‑year CAGR.
   – List at least 3 macro trends and 2 technology/behaviour shifts affecting growth.

2. **Competitive & Benchmark Analysis**
   – Map direct, indirect and substitute competitors.
   – Profile the top 3‑5 best performers: business model, pricing, metrics, recent strategic moves, differentiation.

3. **ROI Persona Prioritisation**
   – For each persona: estimate TAM, willingness‑to‑pay, acquisition cost and LTV.  
   – Score personas (1‑5) on Strategic Fit (considering constraints and expertise), Revenue Potential, Speed‑to‑Market, Resource Intensity.
   – Select and justify the highest‑ROI persona quantitatively.

4. **Van Westendorp Pricing**
   – Produce indicative Too Cheap, Bargain, Expensive, Too Expensive thresholds using secondary data or analogues.
   – Recommend optimal price range and positioning (penetration, value, premium).

5. **Persona Deep Dive**
   – Build psychographic & demographic profile with Jobs‑to‑Be‑Done.
   – List the top 7 urgent pain points your product solves.

6. **Platform & Channel Audit**
   – Identify the 10 platforms (online/offline) where the persona congregates; include audience size & engagement rationale.

7. **Messaging & Authority Roadmap**
   – Distil 3 messaging pillars (functional, emotional, social) with 5 example hooks each.
   – Outline a 12‑month authority‑building plan (content formats, cadence, thought leadership, credibility assets).

8. **Market Entry Strategy**
   – Present a phased go‑to‑market roadmap (0‑3 mo quick wins, 3‑12 mo expansion) aligned with resource constraints.
   – Define critical KPIs and success metrics.

9. **Appendices**
   – Detailed data tables and calculations
   – Source list with URLs
   – Assumptions and confidence levels
</Instructions>

<Constraints>
• Cite every quantitative claim with inline numbered references.
• Business‑formal tone, clear and concise; no hype.
• Use descriptive headings and tables.
• If data unavailable, flag explicitly—do not fabricate figures.
</Constraints>

<Output Format>
**Executive Summary** (≤300 words) – Market outlook, target persona, pricing snapshot, GTM overview, key risks.

**Main Body** ← follow Sections 1‑8 above.

**Appendices**
A. Data Tables
B. Sources
C. Assumptions
D. Glossary
</Output Format>

<User Input>
“Generate the McKinsey‑style market intelligence report.”
</Prompt Template>

Answer any clarifying questions it asks:

Once you provide these details, the AI will generate a comprehensive prompt that we’ll use with Deep Research to conduct market research.

Copy and proceed.

Step 2: Conduct the Market Research

Note: The prompt works best with ChatGPT o3.

• Open a new ChatGPT chat session

• select the model as o3

• toggle the Deep Research button

• paste the prompt generated in Step 1 and hit enter

Answer any follow up questions it asks:

This will take a few minutes to complete. Once done, save the market research into a PDF document so you can refer to it later.

Step 3: Identify the best fit persona

The report will analyze several potential customers and tell you which one offers the highest ROI for your situation.

In my AI consulting business case, the analysis showed content marketers were my golden ticket. They scored high on strategic alignment and were easy to acquire relative to their willingness to pay.

Step 4: Identify the pain points of your ideal persona

The report will reveal the most pressing problems your ideal persona faces that your product/service should resolve.

Step 5: Understand the Ideal Messaging

The report will provide detailed messaging guidance designed specifically for your highest-ROI persona.

For the AI consulting business, it suggested messaging around three content pillars:

These messaging frameworks give you precise language to use in your marketing materials, sales conversations, and content creation.

Step 6: Optimize Your Pricing Strategy

The report will provide pricing guidance based on Van Westendorp analysis - identifying specific price thresholds for your offering:

• “Too Cheap” (price that raises quality concerns)

• “Bargain/Good Value” (perceived as inexpensive but acceptable)

• “Premium but Justifiable” (expensive but worth it if value is clear)

• “Too Expensive” (price that eliminates consideration)

For a new consulting business, the report suggested pricing in the range of $500 — $1500 for an end-to-end setup.

Step 7: Implement the Market Entry Strategy

Finally, the report will outline a practical go-to-market plan:

This gives you a clear roadmap with specific steps to take immediately, then a path to sustainable growth.

The Bottom Line

Don’t blow months targeting people who aren’t ready to buy with messages that don't connect.

Deep Research gives you the power to quickly identify your highest-ROI customers and speak directly to their needs. Use this framework to focus your limited resources where they’ll generate the greatest returns, both now and down the road.

The difference between struggling businesses and thriving ones often comes down to this: knowing exactly who to serve and how to reach them.

Now you have the framework to figure it out.

It’s frustrating.

You just spent hours with AI creating an article but it still sounds robotic.

No personality. Big words that serve no purpose. Mechanical flow.

Before I show you how to fix this, try to identify which of these was written by AI:

In today’s fast-paced digital age, the rise of remote work has significantly transformed the traditional workplace paradigm. On one hand, remote work offers numerous benefits such as increased flexibility and elimination of commuting time. On the other hand, it presents challenges including potential isolation and blurred work-life boundaries.

or

I’ll never go back to a traditional office.

Remote work changed everything for me—and I’m not alone. But let’s be honest about what we’re trading.

The freedom is incredible. I structure my day around when my brain actually works, not when some arbitrary schedule says I should be productive. My output has nearly doubled since ditching my commute. Those two hours in Atlanta traffic? Now they’re my most creative writing time.

Well, both of these are written by AI, but the second one feels human.

By the end of this article, you’ll have a simple system to transform robotic AI content into natural writing that connects with your readers, without spending hours editing.

Subscribe now

Identify the Most Common AI Patterns

The first step is recognizing the obvious signs of AI writing.

These patterns instantly signal to readers (and AI detectors) that your content wasn’t written by a human:

Pattern #1: Overused AI vocabulary and phrases

AI models love certain words and phrases that rarely appear in natural human writing:

delve
dive
embark
groundbreaking
harness
in today’s digital age...
in today’s fast-paced world...
a stark reminder...

Seriously, who uses “embark on a journey” in normal conversation? No one.

Yet AI can’t resist these phrases.

Pattern #2: Excessive transition words

AI relies heavily on explicit transitions that make writing feel mechanical:

firstly
furthermore
however
in contrast
in conclusion

These transitions give the writing an overly structured, academic feel. Natural writing flows without announcing every shift in thought.

Pattern #3: Hedging language overuse

AI avoids strong claims by constantly hedging:

may
might
could
potentially
arguably
seems to
appears to

The excessive caution makes the writing feel uncertain and uncommitted.

Pattern #4: Predictable Paragraph Structures

AI follows consistent paragraph formulas:

• Topic sentence stating the main point

• 2-3 supporting sentences with evidence

• Concluding sentence that restates or transitions

Human writing is more varied, with unexpected structures, single-sentence paragraphs, and less formulaic progression.

Example:

Effective time management is essential for productivity. Studies show that individuals who practice time blocking complete 23% more tasks per day. Additionally, the Pomodoro technique has been proven to improve focus and reduce burnout. Therefore, implementing structured time management techniques will significantly enhance your daily output.

Notice how mechanically it moves from claim to evidence to conclusion.

The Chief Editor Prompt: Your AI-to-Human Conversion Tool

Now that you can spot common AI patterns, here’s the prompt that transforms robotic content into natural writing:

<role>
You are an experienced online editor with a personal, first-person writing style. You write with directness, humor, and occasional personal anecdotes.
You aim to draw readers in with a short, punchy sentence and then keep them engaged with varied sentence lengths and paragraph structures. You take a stand confidently and avoid sounding robotic or formulaic.
</role>

<task>
Your task is to transform the given text into a more personal narrative with a casual, first-person voice.
You will be provided with a file named “AI-typical words and phrases.pdf,” which lists overused words, stock phrases, and transitions to replace or remove.
The goal is to rewrite any text to appear more “human,” with personal touches, varying paragraph structures, and direct language, while preserving the original intent.
</task>

<instructions>
1. Adopt a first-person, conversational tone. Write from your point of view.
Feel free to use casual language or urban slang if it feels natural.

2. Open with a punchy one-liner declarative sentence. Make the very first sentence short (fewer than 10 words) to hook the reader. Keep this as a one line paragraph.
The goal is to lead the reader into the second paragraph.

3. Remove or replace overused words and phrases such as “embark,” “in today’s digital age,” “firstly,” “furthermore,” “however,” “may,” “might,” “could,” “potentially,” “arguably,” “seems to,” “appears to,” etc.
Use natural transitions instead of forced words like “firstly,” “in contrast,” or “in conclusion.”

4. Vary sentence lengths. Have a high perplexity and burstiness.

5. Don't hedge. If something is truly uncertain, state it plainly; otherwise, be direct.
Don’t be afraid to include humor or personal notes.

6. Don't use a strict topic-support-conclusion pattern.
Insert single-sentence paragraphs to emphasize important points or transitions. A paragraph can have a maximum of 3 sentences.

7. Use active voice and direct, assertive statements unless uncertainty is genuinely part of the content.

8. Preserve Facts and Core Intent
Retain the text’s original meaning, data, or arguments.
Do not add new factual information without confirmation.
</instructions>

<constraints>
* Do not introduce new concepts or references beyond what the text originally states.
* Keep the writing lively and direct, avoiding passive or roundabout phrasing.
* Maintain clarity; if a passage could lose meaning, handle it carefully or provide a short note.
* Consult only the “AI-typical words and phrases.pdf” for guidance on overused terms.
</constraints>

<output_format>
Return only the fully revised text.
Do not include commentary about what you changed or why; simply present the final, humanized version.
</output_format>

Are you ready?

To make this system more effective, I have created a file containing a list of the most overused AI words and phrases. Download the PDF here

Once the language model says it’s ready, upload your draft and the downloaded PDF into the same chat thread. The AI will transform your content while keeping the core information intact.

Before and After: See the Transformation

Before:

In today’s fast-paced digital age, the rise of remote work has significantly transformed the traditional workplace paradigm. On one hand, remote work offers numerous benefits such as increased flexibility and elimination of commuting time. On the other hand, it presents challenges including potential isolation and blurred work-life boundaries.

Remote work flexibility allows employees to create schedules that align with their productivity patterns. Studies indicate that 77% of remote workers report higher productivity when working from home. Additionally, the elimination of commuting saves time and reduces stress for many professionals…

After:

I never thought I’d miss office small talk until it vanished.

Remote work hit the workplace like a hurricane, leaving us with a bizarre new reality where pants are optional and your cat is now your closest colleague. I’ve watched this transformation from both sides, and let me tell you—it’s complicated.

Freedom feels amazing.

I create my schedule around when my brain actually functions, not when some arbitrary clock says I should be productive. And I’m not alone here—77% of us remote warriors report cranking out more work from our home bases than we ever did under fluorescent lighting.

No more commuting either. Just think about that extra hour of sleep instead of being sandwiched between strangers on public transport or sitting in traffic contemplating your life choices.

But there’s a dark side…

The Three-Step System for Natural-Sounding Content

Here’s how to implement this system in your workflow:

Step 1: Create the Initial Draft with AI

Start with your favourite AI tool to generate the basic content structure and information.

Step 2: Identify the AI Patterns

Review the draft looking specifically for the four patterns we discussed:

• AI vocabulary and stock phrases

• Excessive transitions

• Hedging language

• Predictable paragraph structures

Step 3: Transform with the Chief Editor Prompt

Copy your draft and the Chief Editor prompt into a chat with Claude or ChatGPT and let it work its magic while you sip on your coffee.

Your inbox is a goldmine going to waste.

You subscribe to dozens of newsletters in your niche for inspiration, but most sit unread in your inbox or get skimmed once and forgotten.

What if you had a system that automatically:

1. Monitored your Gmail inbox for newsletters from your favourite senders

2. Used AI to find insights from those newsletters

3. Generated content ideas inspired by the incoming newsletter but tailored specifically to your audience

4. Provided each article a ‘Potential’ score indicating which ones will likely resonate most strongly

5. Stored everything in Google Sheets so you can instantly find most relevant content ideas for your audience

In this series, we’re building this system.

Here’s a preview of what this system will do

1. Read an unread email from your inbox

2. Push that through the automation

   

3. Output 3 article ideas based on this email but tailored to YOUR audience

   

4. Mark the email as read so we don’t process it again

   

Let’s go!!!

Series Breakdown

Since the whole automation requires quite a few steps, I’ve broken this series into 5 parts:

1. Setting up our Google Cloud Project so that we can automatically monitor our inbox

2. Connecting Gmail to Make, our automation engine

3. Getting an API key from OpenAI

4. Connecting Gmail to ChatGPT and extracting article ideas

5. Storing the article ideas in Google Sheets

For those just joining in, here are parts #1, #2, #3, and #4:

In this final article, we’ll automatically store these AI-generated ideas in Google Sheets, creating a neatly organized content pipeline ready for action.

Note: Nobody will be able to access your inbox automatically apart from you.

This article is more comprehensive than usual and includes many screenshots to guide you through each step. For the best experience, view it on a larger screen.

If you’re reading this in email, it may be clipped due to length. Click here to view the full tutorial on the website.

Let’s begin.

Subscribe now

Step 1: Add the Iterator Module

1. From your Make dashboard, open the scenario you created in the previous articles (the one that already has Gmail and OpenAI).

2. Click the ＋ icon after your OpenAI module to add a new module

3. Search for “Iterator” and select it. This module will fetch the content ideas from ChatGPT one by one.

4. Drag content_ideas[] (generated by the OpenAI module) into the Array field in the Iterator module.

Step 2: Set Up Your Google Sheet

• In your Google Drive, create a new folder named “Automatic Content Ideation”.

• Inside that folder, create a new Google Sheet named “Article Ideas”. We will store the content ideas in this sheet.

Create the following five columns in that sheet:

• Title

• Description

• Pain Point Resolved

• Promise to the reader

• Potential

Step 3: Connect the Iterator Module to Google Sheets

1. From your Iterator module, click the ＋ icon to add another module.

2. Search for Google Sheets and select “Add a Row”.

Step 4: Configure Google Sheets Module

• Set Search Method to “Search by path”.

• Under Drive, choose “My Drive”.

• Click on “Click here to choose file” under Spreadsheet ID.

  

• Select the “Automatic Content Ideation” folder

  

• Click on ＋ next to the folder name and choose your “Article Ideas” Sheet.

  

  

• Choose the Sheet Name (e.g., Sheet1) where you want to store the article ideas.

  

• Under “Table contains headers”, select “Yes” (because the first row in your Google Sheet is column names).

• Map each field from the Iterator module to the matching columns:

  • Title (A) ← title (from the Iterator)

  • Description (B) ← description

  • Pain Point Resolved (C) ← pain_point_resolved

  • Promise to the reader (D) ← promise_made

  • Potential (E) ← potential

• Click “Save.”

Step 5: Test Your Automation

• Right-click on the Gmail module and select “Choose where to start”

• Select “All emails”, and click on Save

• Make sure you have at least one unread newsletter email in the label you’re monitoring.

• Click “Run once”. Wait a few seconds for the automation to run.

• Check your “Article Ideas” Google Sheet. You should see new rows with the content ideas generated by ChatGPT.

  Here’s the email I processed:

  

• It was an opinion piece by Ethan Mollick on which AI to use. Our system took the key ideas from this email and created these three article ideas

According to ChatGPT, the first article ideas: “Eliminating Filler Content” will resonate the most with my audience (Potential: 8.8/10)

Step 6: Schedule Your Automation

• Right-click on the Gmail module and select “Choose where to start”

  

• Select “All emails”, and click on Save

• Click the scheduler at the bottom and select an appropriate interval when you want the automation to run.

Congratulations!

That’s it! You now have a fully automated system that:

1. Fetches unread newsletters from Gmail.

2. Analyzes them with ChatGPT.

3. Iterates through each AI-generated idea.

4. Stores each idea in your “Article Ideas” Google Sheet.

So, now, instead of scratching your head to come up with article ideas, you have a database of readymades ideas with their potential for success.

All you need to do now is research and create the article.

Enjoy your new streamlined system!

Want personalized help building your customized AI system?

I offer 90-minute sessions where we build your custom AI content system together.

Learn More About 1:1 Sessions

Join the Deep Writing Community

Your newsletter subscriptions contain valuable insights that could fuel your content creation for months.

But manually extracting these ideas is time-consuming and inconsistent.

So we’re using AI to analyze newsletters, identify the most relevant content for our audience and generate fresh article ideas tailored to them.

This system will:

1. Monitor your Gmail inbox for newsletters from your favourite senders

2. Use AI to find insights from those newsletters

3. Generate content ideas inspired by the incoming newsletter but tailored specifically to your audience

4. Provide each article a ‘Potential’ score indicating which ones will likely resonate most strongly

5. Store everything in Google Sheets so you can instantly find most relevant content ideas for your audience

All without you lifting a finger.

Here’s a preview of what this system will do

1. Read an unread email from your inbox

2. Push that through the automation

3. Output 3 article ideas based on this email but tailored to YOUR audience

Let’s go!!!

Series Breakdown

Since the whole automation requires quite a few steps, I’ve broken this series into 5 parts:

1. Setting up our Google Cloud Project so we can automatically monitor our inbox

2. Connecting Gmail to Make, our automation engine

3. Getting an API key from OpenAI

4. Connecting Gmail to ChatGPT and extracting article ideas

5. Storing the article ideas in Google Sheets

For those just joining in, here are parts #1, #2, and #3:

Today, we’ll connect ChatGPT to Gmail through the API, so it can analyze incoming newsletters and generate relevant ideas customized for our audience.

Note: Nobody will be able to access your inbox automatically apart from you.

This article is more comprehensive than usual and includes many screenshots to guide you through each step. For the best experience, view it on a larger screen.

If you’re reading this in email, it may be clipped due to length. Click here to view the full tutorial on the website.

Let’s begin.

Subscribe now

1. Preparing Your Publication’s Core Information

Before we integrate ChatGPT, you’ll need to gather some essential information about your publication.

This will help ChatGPT understand what’s relevant to your audience and generate truly valuable content ideas.

Take some time to define these four key elements (I’ve linked to detailed guides for each):

Value Statement

What core value does your publication provide? This is the overarching purpose that explains why readers should care.

Here’s how to craft a compelling value statement using AI

Ideal Reader Pain Points

What specific challenges does your audience face?

Here’s how to identify your ideal reader’s pain points using AI

Movement Statement

What change are you trying to create? This is your rallying cry or mission statement.

Here’s how to craft a movement statement with AI that resonates

Unique Value Proposition (UVP)

What makes your content uniquely valuable? This describes what sets your content apart from others in your space.

Here’s how to craft a UVP using AI

If you already have these elements defined for your publication, great! If not, the linked guides will walk you through creating each one with AI.

Keep these handy as we’ll need them in Step 4 to customize the ChatGPT prompt.

2. Adding OpenAI to Your Make Scenario

First, let’s connect Make to OpenAI so it can access ChatGPT:

• From your Make dashboard, open the scenario we created in the previous article

• Click the ＋ icon after your Gmail module to add a new module

• In the search box, type “OpenAI” and click Show more to see all OpenAI options

• Select Create a Chat Completion

3. Setting Up Your OpenAI Connection

• Click Create a connection to link your OpenAI account

• Give your connection a name

• Paste your OpenAI API key you generated in the previous article

• Click Save to store your connection

4. Tailoring the Prompt Behind Your System’s Intelligence

Here’s the template you’ll use to instruct ChatGPT on how to analyze newsletters and generate content ideas.

I’ve designed it with placeholders in the format [XYZ] that you’ll need to replace with your own information:

<System>
You are an expert content ideator at [PUBLICATION NAME] ([PUBLICATION ABBREVIATION]) - a [NEWSLETTER/BLOG].

You specialize in coming up with article ideas that establishes [PUBLICATION ABBREVIATION] as an authority in our niche and resonates with the target audience.
</System>

<Context>
The goal is to come up with new content ideas based on incoming newsletters.

Below are the details about [PUBLICATION ABBREVIATION]:
#####
**Value Statement** (overarching purpose and value of [PUBLICATION ABBREVIATION])
[VALUE STATEMENT]
#####
**Ideal Reader Pain Points** (main challenges faced by the target audience of [PUBLICATION ABBREVIATION])
[PAIN POINTS]
#####
**Movement Statement** (rallying cry or mission driving [PUBLICATION ABBREVIATION])
[MOVEMENT STATEMENT]
######
**Unique Value Proposition (UVP)** (what makes the content of [PUBLICATION ABBREVIATION] uniquely valuable)
[UVP]
#####

The ideas you generate will be used to create content with the goal to educate and establish authority.
</Context>

<Task>
1. Review the provided newsletter text to understand its main topic, tone, and focus.

2. Generate 10 content ideas around the main idea of the newsletter. While generating each content idea, ensure that: it solves a very narrow problem (the narrower the better), and provides a tangible outcome. Break an idea down till it can't be broken further.

3. Evaluate each generated content idea on these four parameters (weightage of each is provided in square brackets):
* Atomicity: How atomic/small is the problem we're solving? (Smaller the better) [40%]

For example,

- "How to use AI for writing" is a very broad topic and will get the lowest score in atomicity.
- "How to use AI to conduct research for content creation" is better and more tangible.
- "How to use AI to conduct research and find statistics relevant to your article" is still better because it focuses on a narrower problem.
- "How to use Google Deep Research to conduct research and find statistics relevant to your article" is the best as it's the narrowest of the four.

* Impact: How strongly does this resonate with the ideal reader's pain points? [20%]
* UVP alignment: Does it complement the overarching reason [PUBLICATION ABBREVIATION] exists? [15%]
* Movement alignment: Does this content support or inspire the mission of [PUBLICATION ABBREVIATION]? Example: If the movement is about helping creators monetize, does this article’s content reinforce that concept? [15%]
* Timeliness: How relevant is this in the current moment or near future for the target audience? [10%]

4. Suggest top 3 content ideas with the highest scores. Include a concise, 3–5 sentence description for each idea.

5. For each of the three content ideas:
- mention the pain point(s) we are solving
- promise we will be making in the article to the reader
</Task>

<Constraints>
- Provide objective scores (0–10) for each parameter without bias.
- Keep responses concise; do not exceed relevant details.
- Ensure that each content idea solves a specific, narrow problem for the target audience.
- Always return the response strictly in valid JSON format.
- Do not include any explanatory text, preamble, or formatting outside of JSON.
- Ensure the output follows the structure exactly as specified in the `<Output Format>` section.
</Constraints>

<Output Format>
Return the response strictly in the following JSON format:
```
{
  "content_ideas": [
    {
      "title": "[Idea 1 Title]",
      "description": "[3-5 sentence description]",
      "pain_point_resolved": "[The pain point(s) this article will resolve]",
      "promise_made": "[What transformation will the reader experience? What is the promise we will make to the reader?]",
      "potential": "[Overall Relevance Score (0-10)]"
    },
    {
      "title": "[Idea 2 Title]",
      "description": "[3-5 sentence description]",
      "pain_point_resolved": "[The pain point(s) this article will resolve]",
      "promise_made": "[What transformation will the reader experience? What is the promise we will make to the reader?]",
      "potential": "[Overall Relevance Score (0-10)]"
    },
    {
      "title": "[Idea 3 Title]",
      "description": "[3-5 sentence description]",
      "pain_point_resolved": "[The pain point(s) this article will resolve]",
      "promise_made": "[What transformation will the reader experience? What is the promise we will make to the reader?]",
      "potential": "[Overall Relevance Score (0-10)]"
    }
  ]
}
```

Here’s how to customize the prompt template:

• Replace [PUBLICATION NAME] with your newsletter or blog name

• Replace [PUBLICAION ABBREVIATION] with a short version of your publication name (appears multiple times)

• Choose between [NEWSLETTER/BLOG] based on your publication type

• [VALUE STATEMENT]: Your publication’s value statement

• [PAIN POINTS]: Your ideal reader’s pain points

• [MOVEMENT STATEMENT]: The change you’re trying to create through your publication

• [UVP]: Your publication’s UVP

In this prompt, we’re telling ChatGPT to:

• Act as an expert content ideator and analyze the incoming newsletters to understand its main topic, tone, and focus.

• Generate 10 content ideas around the main idea of the newsletter, each solving a narrow, targeted problem.

• Rate each of the 10 content ideas on the following 5 parameters (0-10):

  • Atomicity: How narrow is the problem this article will solve. To ensure that we comprehensively address a problem in our article, it’s important to choose a narrow problem. Generally, the narrower, the better.

  • Impact: How strongly does this resonate with our audience’s needs, pain points, and goals?

  • UVP alignment: How strongly does it complement the overarching reason our content exists?

  • Movement alignment: How strongly does this content support or inspire our mission?

  • Timeliness: How relevant is this content in the current moment or near future for our audience?

• Assign different weightages for each of the parameters (40% for atomicity, 20% for resonance with ideal reader’s pain points, 15% each for UVP and movement alignment, 10% for timeliness).

• Combine the 5 scores to find an overall impact score (0-10). Higher the score, more relevant the article idea is to your target audience.

• Return the top 3 content ideas based on the overall impact score.

5. Configuring the Intelligence

Now we’ll set up how ChatGPT will analyze your newsletters.

• Under Select Method, select Create a Chat Completion from the dropdown

• For Model, choose o1 (system)
  This is OpenAI’s reasoning model that takes time to “think” before responding. This is perfect for analyzing whether an incoming newsletter is relevant to our audience thoughtfully rather than rushing to conclusions.

• Under Messages, click Add message

• Select Developer/System as the role. This first message is essentially your “instructions” to ChatGPT. It’s like hiring an expert and explaining what you want them to do.

• Paste the prompt created in Step #4 into Text Content field

6. Feeding the Newsletter Content Into ChatGPT

Now we need to tell ChatGPT to analyze the newsletter content.

• Click Add message again to create a second message

• This time, select User as the role

• Click in the Text Content field

• You’ll see a panel with red tags appear. Find the tag that says “Text Content” and drag it to this field. This feeds the actual email content from the Gmail module to ChatGPT.

7. Configuring Output Settings

• Toggle on Show advanced settings at the bottom

• Set Response Format to “JSON Object”

• Set Parse JSON Response to “Yes”

• Set Reasoning effort to “Medium”

• Click Save to complete the module setup

Let’s Bring Your System to Life

It’s time to test if your newsletter analyzer works.

• Make sure you have at least one unread newsletter email in the label you are monitoring

• From your Make dashboard, click Run once to process that email

If the automation doesn’t run beyond Gmail, click on the bubble above to see what’s the issue.

If you see an error saying “The module didn’t return any new data,” select Choose where to start → From now on → Save, and try running again.

• When the run completes, click on the OpenAI bubble to see the results

• Click the ＋ icons to expand different sections of the output

• Expand Result to see what ChatGPT has returned.

• You’ll see a content_ideas tag. This contains the content ideas

• Expand 1 under content_ideas to see the first content idea, 2 for the second, and 3 for the third.

• Each content idea has a title, description, the pain point it will solve for YOUR audience, the promise we’ll make in the article, and the potential of the article for your audience.

What You’ve Accomplished & What’s Next

Congratulations! You’ve now successfully:

1. Connected Gmail to Make.com to monitor your newsletter subscriptions

2. Set up ChatGPT to analyze incoming newsletters

3. Created a system that generates tailored content ideas relevant to your audience

4. Tested the workflow to ensure everything functions correctly

At this point, your system can evaluate newsletters and generate content ideas, but there’s one final piece to make this truly valuable: storing these ideas for future reference.

In the next article, we’ll complete our automation by connecting the OpenAI module to Google Sheets and storing the generated content ideas and relevance scores.

Want personalized help building your customized AI system?

I offer 90-minute sessions where we build your custom AI content system together.

Learn More About 1:1 Sessions

Join the Deep Writing Community

You have the expertise and you know your stuff.

But turning that knowledge into consistent, authority-building content can be a challenge.

You’re spending 5-8 hours per piece, starting from scratch each time, wrestling with AI tools.

Your publishing schedule is reactive. Your valuable knowledge stays trapped in your head or scattered across notes. And content creation feels like a second full-time job.

There’s a better way.

What This Is

A focused 90-minute implementation session where we build your custom AI content system together—specifically designed for your expertise and audience.

This isn’t generic AI advice or theoretical frameworks.

We’ll create tailored workflows that capture your unique voice, leverage your accumulated knowledge, and turn content creation from a reactive scramble into a systematic process.

You’ll walk away with working systems, not just ideas.

The Process

1. Pre-Session Assessment: Before we meet, I’ll analyze your content style, expertise areas, and biggest bottlenecks to understand exactly what systems you need.

2. 90-Minute System Building Session We’ll build your custom AI workflows together, focusing on:

   1. Converting your expertise into systematic content ideation

   2. Creating prompts that maintain your voice and industry insights

   3. Building workflows that connect research → creation → optimization

3. Custom System Documentation You’ll receive 2-3 tailored workflows designed specifically for your expertise and content goals, plus implementation templates.

4. Strategic Implementation Blueprint A step-by-step plan to integrate these systems into your workflow, including how to scale them as your authority grows.

5. Two Weeks of Implementation Support Email support as you implement your new systems, plus troubleshooting for any integration challenges.

6. Complete Session Recording Reference our entire strategy and system-building session anytime.

Results You Can Expect

• Transform overwhelming long-form projects into manageable, systematic processes

• Significantly reduce time spent on research and first drafts

• Improve content quality by leveraging AI for deeper research and insights

• Scale your business through content without burnout or overwhelm

This Is Perfect For You If:

• You create long-form content (newsletters, blogs, articles) and want to scale output without sacrificing quality

• You want to leverage AI tools while maintaining your unique voice

• You’re struggling with consistent publishing schedules

• You’re a freelance writer looking to increase output to boost income

• You’re a writing entrepreneur wanting to free up time for strategy and growth

This Is NOT For You If:

• You’re primarily focused on book writing or creating video content

• You need help with specific writing software beyond ChatGPT/Claude/Gemini

• You’re looking for someone to write content for you

Investment: $297

100% Satisfaction Guarantee: If you implement the system as instructed and don’t experience at least a 30% improvement in your content creation efficiency within 14 days, I’ll issue a full refund of your payment minus the $15 transaction processing fee.

Limited Availability

These sessions require my full attention and customization, so I can only offer 8 spots, with only 1 consultation per day.

Ready to Transform Your Content Process?

Book Your Consultation Now

Your newsletter subscriptions are a goldmine of content ideas.

But keeping up with dozens of newsletters every week and extracting the best insights can be challenging for busy writerpreneurs.

What if those newsletters could automatically transform into content ideas while you sleep?

In this series, we’re building an AI-powered system that:

1. Monitors your Gmail inbox for newsletters from your favourite senders

2. Uses AI to find insights from those newsletters

3. Generates content ideas inspired by the incoming newsletter but tailored specifically to your audience

4. Provides each article a ‘Potential’ score indicating which ones will likely resonate most strongly

5. Stores everything in Google Sheets so you can instantly find most relevant content ideas for your audience

All without you lifting a finger.

Here’s a preview of what this system will do

1. Read an unread email from your inbox

2. Push that through the automation

3. Output 3 article ideas based on this email but tailored to YOUR audience

   

Let’s go!!!

Series Breakdown

Since the whole automation requires quite a few steps, I’ve broken this series into 5 parts:

1. Setting up our Google Cloud Project so that we can automatically monitor our inbox

2. Connecting Gmail to Make, our automation engine

3. Getting an API key from OpenAI

4. Connecting Gmail to ChatGPT and extracting article ideas

5. Storing the article ideas in Google Sheets

Here are parts #1 and #2:

Now it’s time to add the brain to our operation: connecting ChatGPT’s AI through the OpenAI API to transform those newsletters into actionable content ideas.

Note: Nobody will be able to access your inbox automatically apart from you.

This article is more comprehensive than usual and includes many screenshots to guide you through each step. For the best experience, view it on a larger screen.

If you’re reading this in email, it may be clipped due to length. Click here to view the full tutorial on the website.

Let’s begin.

Subscribe now

What is an API Key and Why Do You Need One?

An Application Programming Interface (API) acts as a bridge that enables different software programs to communicate with each other.

Normally, when we interact with ChatGPT, we manually type in the commands and read the responses.

This is okay for one off conversations.

But, if we have to analyze 20 different newsletters to identify whether they are relevant to our niche, and then extract potential content ideas, the manual method is too cumbersome.

We can automate this using an API where we can:

• Send the newsletter content to ChatGPT via an API

• ChatGPT sends the response back through the API

Using APIs offers several advantages over manual approaches:

• Automation: Process newsletters 24/7 without your involvement

• Scale: Handle dozens or hundreds of newsletters efficiently

• Consistency: Apply the same analytical criteria to each newsletter

• Speed: Process content in seconds rather than hours

To communicate automatically with ChatGPT, we’ll need an API key.

API keys function like secure passwords that allow different services to communicate programmatically. They’re how we’ll connect Make.com to ChatGPT.

Let’s get your OpenAI API key set up.

Important Note: API usage is billed separately from any ChatGPT Plus subscription you might have.

You’ll need to add a payment method to your OpenAI account, and you’ll be charged based on your API usage.

For the system we’re building, costs are typically minimal (~$5 per month for moderate usage).

1. Create an OpenAI Account

• Go to OpenAI’s platform and sign in or create an account

2. Set Up a Project in OpenAI

• Once you log in, you’ll be directed to the home page

• Click on Create a new project

  

• Give your project a meaningful name

• Click Create

Projects help you organize and track your API usage, which is essential for managing costs and monitoring your automation system’s performance.

3. Create Your API Key

With your project set up, it’s time to create the actual API key.

• In your project dashboard, find and click on API Keys in the left sidebar

• Click the Create new secret key button

• Give your key a name

• Click Create secret key

4. Save Your API Key Securely

This is the most important step.

When your API key is generated, you’ll see it displayed only ONCE.

• Copy your API key immediately (it starts with “sk-”)

• Store it securely

• NEVER share it publicly

⚠️ Warning: If you lose your API key, you cannot recover it—you’ll need to create a new one.

Anyone with your API key can use OpenAI services at your expense, so treat it like your credit card number!

5. Set Up Billing

To use the OpenAI API, you’ll need to add a payment method:

• Navigate to Billing in the left sidebar

• Click Add to credit balance

• Enter a starting amount ($5-10 is sufficient for our initial testing)

• Add your payment details and complete the transaction

API usage is charged based on the number of tokens processed (roughly ~750 words equals 1000 tokens).

For our newsletter processing system, costs will be minimal—typically less than $5 per month for moderate usage.

This is completely separate from any ChatGPT Plus subscription you might have, which only covers web-based usage of ChatGPT.

6. Set Usage Limits (Highly Recommended)

To prevent unexpected charges, I strongly recommend setting usage limits:

• Go to Usage limits under the Billing section

• Set a monthly spending cap (e.g., $20)

• Enable email notifications when you reach a certain threshold (e.g., $10)

• Click Save

This step is crucial for peace of mind.

It ensures your automation can run continuously without risk of unexpected costs.

Keeping an eye on your API usage and balance

As you work on your automation, I strongly urge you to keep an eye on your API usage and balance.

• To check API usage, go to the projects Dashboard → Usage

• To check your credit balance, go to Settings → Billing

Congratulations! Your OpenAI API key is now ready to power your automated content idea system.

Ready to move forward?

Want personalized help building your customized AI system?

I offer 90-minute sessions where we build your custom AI content system together.

Learn More About 1:1 Sessions

Join the Deep Writing Community

Your newsletter subscriptions are a goldmine of content ideas.

Each one contains ideas that could inspire your next viral post or solve a content creation challenge.

But with dozens arriving weekly, it’s impossible to extract every valuable insight manually.

What if those newsletters could automatically transform into content ideas that your audience would love while you sleep?

In this series, we’re building an AI-powered system that:

1. Monitors your Gmail inbox for newsletters from your favourite senders

2. Uses AI to find insights from those newsletters

3. Generates content ideas inspired by the incoming newsletter but tailored specifically to your audience

4. Provides each article a ‘Potential’ score indicating which ones will likely resonate most strongly

5. Stores everything in Google Sheets so you can instantly find most relevant content ideas for your audience

All without you lifting a finger.

Here’s a preview of what this system will do

1. Read an unread email from your inbox

2. Push that through the automation

3. Output 3 article ideas based on this email but tailored to YOUR audience

   

4. Mark the email as read so we don’t process it again

Let’s go!!!

Series Breakdown

Since the whole automation requires quite a few steps, I’ve broken this series into 5 parts:

1. Setting up our Google Cloud Project so that we can automatically monitor our inbox

2. Connecting Gmail to Make, our automation engine

3. Getting an API key from OpenAI

4. Connecting Gmail to ChatGPT and extracting article ideas

5. Storing the article ideas in Google Sheets

We’ve already completed part #1 here:

Today, we’ll connect our Gmail account to Make so we can automatically monitor our inbox through Make.com for newsletters from our favorite senders.

Note: Nobody will be able to access your inbox automatically apart from you.

This article is more comprehensive than usual and includes many screenshots to guide you through each step. For the best experience, view it on a larger screen.

If you’re reading this in email, it may be clipped due to length. Click here to view the full tutorial on the website.

Let’s begin.

Subscribe now

What is Make?

Make.com is a powerful automation platform that connects different apps together without coding.

For our newsletter content extraction system, Make is the engine that powers everything, connecting Gmail, AI tools, and storage solutions.

The free plan allows for 1,000 operations per month, which is more than enough for our newsletter monitoring system.

If you don’t already have a Make account, head over and sign up for their free plan.

1. Create a New Scenario in Make.com

• Log in to Make.com and go to your Dashboard.

• Click on Create a new scenario.

• This will open a blank workflow canvas where we’ll build our automation.

2. Add the Gmail Module

• Click on the “＋” (Add Module) button in the center of the canvas.

  
  This is where we’ll start building our workflow. Think of each module as a building block that performs a specific function.

• Search for “Gmail” in the module list.

• Select Watch Emails.
  
  The Watch Emails module is our starting point. It will check your Gmail account for new emails based on criteria we set.

3. Connect Gmail to Make.com

This is where the Google Cloud Project we set up in the last article comes into play. We need to use the credentials we created to establish the secure connection between Make.com and Gmail.

• In the configuration window, click Create a connection to create a new Gmail connection.

• In the connection window:

  • Give your connection a name (like “Newsletter Gmail Connection”)

  • Do not click “Sign in with Google” yet. We need to configure some settings first

• Toggle Show advanced settings at the bottom of the screen. You’ll see two fields appear:

  • Client ID

  • Client Secret

• It’s time to use the Google Cloud credentials we saved in the last article.

  • Copy the Client ID from your Google Cloud Project and paste it in the Client ID field

  • Copy the Client Secret from your Google Cloud Project and paste it in the Client Secret field

• In case you didn’t save them, here’s how you can retrieve the credentials:

  • Go to https://console.cloud.google.com/auth/clients

  • Click on the download button to view your credentials

• Now click Sign in with Google.

• You’ll be taken to a Google authorization screen. Select the Gmail account that you added as a test user in the Google Cloud Project.

• You might see a warning that “Google hasn’t verified this app.” This is normal for applications in testing mode. Click Continue.

Don’t worry about the warning. Since you created this Google Cloud Project yourself, you’re essentially connecting to your own app.

• Next, you’ll see a permission screen asking to grant Make.com (which might show as “integromat.com”) access to your Gmail.

Make was previously called "Integromat," so don’t be confused if you see this name during the setup process - they’re the same platform.

• Make sure to check the box that allows it to “Read, compose, send, and permanently delete all your email from Gmail.”

• Click Continue to complete the connection.

Note: If you encounter an error saying “Resource not found,” close the window, go back to Make.com, and restart the connection process.

• When the connection is successful, you’ll see your connection name appear in the dropdown

Congratulations! Your Gmail account is now connected to Make.com.

4. Set Up a Newsletter Label in Gmail (Recommended)

You have two options for monitoring newsletters:

1. Monitor your entire inbox (simpler but less focused)

2. Monitor only specific newsletters using labels (recommended)

I recommend using labels to focus only on high-quality newsletters relevant to your niche. This prevents your system from getting cluttered with promotional emails or personal messages.

If you prefer to monitor your entire inbox, you can skip to the next section.

To set up a dedicated newsletter label:

• Open your Gmail account in a new tab.

• Find a newsletter you want to monitor and open it.

• Click on the three dots at the top of the email.

• Select Label as → Create New from the dropdown

• Name your label “Newsletters to Monitor” or “Automatic Monitoring” and click Create.

• Copy the email ID of the newsletter you want to monitor

• Click on the three dots at the right side of the email and select Filter messages like this

• In the window that pops up, mention the email ID you just copied into the From field

• Click Search

• Next, tick the checkbox that says “Apply the label” and select the label you just created from the dropdown

• Also, select the checkbox that says “Also apply filter to …” so that all the previous emails from this sender also gets labelled

• Click Create Filter

Repeat this process for other newsletters you want to include in your monitoring system.

This selective approach gives you complete control over which sources feed into your content idea system.

5. Configure the Gmail Module

Now, let’s configure how the Gmail module created in Make monitors your inbox:

• For Folder, you have two options:

  • If using labels (recommended): Click on “Click here to choose a folder” and select the label you just created

  • If monitoring your entire inbox: select INBOX

• Set Filter type to “Simple filter.”

• For Criteria select “Only Unread emails” so we do not process the same emails repeatedly

• Set Mark email message(s) as read when fetched to “Yes” - this ensures we don’t process the same email twice.

• For Maximum number of results, enter “1” for testing. We’ll increase this later when everything is working properly.

• Click Save to apply these settings.

6. Test Your Scenario

Let’s make sure your connection is working correctly:

• Once done, Make will take you back to the dashboard. Click on “Select the first email” and it should retrieve the first unread email

• Click on the fetched email. A tick mark should appear

• Click Save

• Click on Run once from the Make dashboard

• Make will check your Gmail account for the most recent unread email that matches your criteria (either from your label or your inbox).

• It will then display the details of the email in a popup window and mark the email as read

• To view the text of the email, expand the “Text content” section in the results.

Boom! You can now automatically monitor your inbox from Make.

Why This Matters for Your Writing Business

This automation fundamentally changes how you mine other newsletters for content ideas.

Instead of spending hours manually sifting through your inbox hoping for inspiration, you’re building a system that works for you 24/7.

The Gmail label approach gives you complete control over which newsletters feed into your system.

You can:

• Include only high-quality, relevant newsletters

• Easily add or remove sources from your monitoring

• Keep your inbox organized while the automation works in the background

For writing entrepreneurs, this means:

• Never missing valuable ideas hidden in your inbox

• Reclaiming hours of research time every week

• Building a sustainable content creation system that scales with your business

• Creating a permanent asset that grows more valuable over time

Ready to move forward?

Want personalized help building your customized AI system?

I offer 90-minute sessions where we build your custom AI content system together.

Learn More About 1:1 Sessions

Join the Deep Writing Community

Your inbox is a goldmine going to waste.

You subscribe to dozens of newsletters in your niche for inspiration, but most sit unread in your inbox or get skimmed once and forgotten.

What if you had a system that automatically:

1. Monitored your Gmail inbox for newsletters from your favourite senders

2. Used AI to find insights from those newsletters

3. Generated content ideas inspired by the incoming newsletter but tailored specifically to your audience

4. Provided each article a ‘Potential’ score indicating which ones will likely resonate most strongly

5. Stored everything in Google Sheets so you can instantly find most relevant content ideas for your audience

In this series, we’ll build this system.

Here’s a preview of what this system will do

1. Read an unread email from your inbox

2. Push that through the automation

3. Output 3 article ideas based on this email but tailored to YOUR audience

   

4. Mark the email as read so we don’t process it again

Let’s go!!!

Series Breakdown

Since the whole automation requires quite a few steps, I’ve broken this series into 5 parts:

1. Setting up our Google Cloud Project so that we can automatically monitor our inbox

2. Connecting Gmail to Make, our automation engine

3. Getting an API key from OpenAI

4. Connecting Gmail to ChatGPT and extracting article ideas

5. Storing the article ideas in Google Sheets

While this setup requires some initial effort, it creates a permanent asset for your writing business — an ever-growing database of content ideas curated specifically for your audience.

Note: Nobody will be able to access your inbox automatically apart from you.

This article is more comprehensive than usual and includes many screenshots to guide you through each step. For the best experience, view it on a larger screen.

If you’re reading this in email, it may be clipped due to length. Click here to view the full tutorial on the website.

Let’s begin.

Subscribe now

What is Make?

Make.com is an automation platform that lets you connect different apps together (like Gmail and AI tools) using a simple visual interface. No coding required.

For this tutorial, we’ll be using Make’s free plan.

Step-by-Step Gmail Connection Process

Here’s what we’ll accomplish in this article:

1. Create a Google Cloud project

2. Enable the Gmail API

3. Set up authentication

4. Add authorized domains

5. Configure test users

6. Set up API permissions

7. Create client credentials

Configuring your Gmail account for Make involves a few technical steps.

But don’t worry.

I’ve broken everything down into manageable pieces. This is a one-time setup that will enable all your future Gmail automations.

Note: Make.com was previously called “Integromat” and some references to the earlier name still appear in the setup process. Don’t get confused — they’re the same platform.

1. Creating Your Google Cloud Project

First, we need to create a project in Google Cloud that will allow Make to access your Gmail account:

• Go to https://console.developers.google.com/ and log in with the Gmail account you want to monitor.

• Even though we’re using the Google Developers site, you won’t need to write any code. Also, this is needs to be done only once for each Gmail account.

• If this is your first time accessing the developers console with this email, you’ll need to agree to Google’s terms and conditions.

• Once you’ve agreed to the terms, you’ll land on the APIs and Services dashboard.

• Click on Create project

• Give your project a name like “Newsletter Automation” or “Gmail-Make Integration” and click Create.

• Wait for Google to finish creating your project. You’ll be taken to the project dashboard once it’s complete.

✅ Progress Check: You’ve created a Google Cloud project that will house your Gmail integration. This is the foundation for your automation system.

2. Enabling the Gmail API

Now we need to give the project permission to work with Gmail:

• From your project dashboard, click on Enable APIs and Services. It will take you to the API Library dashboard.

• In the API Library search bar, search for “Gmail” and click on “Gmail API”

• You’ll see two results: “Gmail API” and “Gmail Postmaster Tools API”

• Select “Gmail API”

• On the Gmail API page, click the Enable button

✅ Progress Check: You’ve enabled the Gmail API, which gives your project the ability to interact with your Gmail account.

3. Setting Up Authentication

After enabling the API, we need to create a consent screen (this is what you’ll see when connecting Make.com later):

• From the sidebar, click on OAuth consent screen

• In the next screen, select Overview from the sidebar, and click on Get Started. This will take you to the Project Configuration screen

• In the Project Configuration screen, you’ll see the App Information section

• In the textbox that says App name, enter “Make Gmail Integration”

• In the textbox that says User support email, select your Gmail address from the dropdown

• Click Next. This will take you to the Audience section

Note: Nobody will be able to use this application except for people you personally authorize as test users later in this process. So don’t worry.

• In the Audience section, you need to specify the user type that will be able to access the project

  • “Internal” - This will allow other users inside your organization to access this project. But, since we’re building an integration for a personal gmail account, this doesn’t apply.

  • “External” - This will allow any test user with a Google Account to access this Project. Later in the process, we will add our own Gmail ID as the test user, and the application will only be available to us.

• Since we’re setting up for a personal Gmail account, select “External”

• Click Next

• In, the Contact Information section, you’ll need to provide an email ID where Google will send relevant updates about your project. Enter your Gmail address

• Click Next

• In the Finish section, check the checkbox to agree to Google API Services policy

• Click Continue and then Create

✅ Progress Check: You’ve set up authentication for your project, which controls who can access your Gmail data through this integration.

You’re halfway through the setup process!

4. Adding Authorized Domains

Next, we need to add the domains that will be allowed to use this connection:

• Select Branding from the sidebar

• Scroll down to Authorized domains

• Click Add domain

• You’ll need to add two domains here:

  make.com

  integromat.com

• Ensure that you add both domains. Make used to be known as Integromat before, and it still uses the old name in the backend

• After adding both domains, click Save

5. Adding Test Users

Now we need to create test users who will be able to access your project:

• Select Audience from the sidebar

• Under Test users, click Add Users

• In this screen, enter your Gmail account that you want to use in Make

• Click Save

✅ Progress Check: You’ve authorized Make.com domains and added yourself as a test user. This ensures that only you can use this connection to access your Gmail.

6. Setting Up API Permissions

In this step, we’ll define the functions we can perform in Gmail through Make:

• Select Data Access from the sidebar

• Click Add or remove scopes

• In the screen that opens, type “Gmail” in the search box and select “Gmail API” from the dropdown

• Set the rows per page to 20

• Select the following scopes by checking their boxes:

  • mail.google.com

  • gmail.modify

  • gmail.compose

  • gmail.readonly

  • gmail.metadata

  • gmail.insert

  • gmail.send

  • gmail.labels

• Click Update at the bottom of the screen

• In the next screen, scroll down to the bottom and click Save

✅ Progress Check: You’ve defined what actions your automation can take on your Gmail account.

7. Creating Client Credentials

We’re in the final step of configuring the project.

Now we need to create credentials that Make will use to connect to your Gmail account:

• Select Clients from the sidebar

• Click Create Client

• In the Application type dropdown, select Web application