I wanted to understand how one specific creator had built his business, and that isn't something you can usually find with a quick search. Once in a while, you'll find a breakdown of somebody's titles, thumbnails, or posting schedule. That's all fine, but it wasn't what I needed. I wanted to know how the whole business worked: how he got attention, who the audience was, what he sold, how the money might work, which tools he used, and who else helped run it. So that's what I built.
What kept getting in the way
I could find plenty of claims. Proof was harder.
While I was doing the research, I kept running into the same problem. Prices, videos, products, and public claims were easy enough to find. Real revenue, conversion rates, expenses, and the tools being used behind the scenes usually weren't. I didn't want to fill those blanks with guesses and then pass them off as facts. So the framework makes you say what you know, what somebody reported, what you're figuring out from the evidence, and what you still don't know.
What you get
The guide, the prompt, and a real example.
The full guide, broken into ten parts so you can work through one piece at a time.
Five labels that make it clear what you know, what somebody claimed, what you're figuring out, what you can use, and what nobody can prove.
A research process that tells you what to look for, where to look, and when to stop guessing.
A way to look at prices and possible revenue without pretending you know what the business actually makes.
A scorecard that helps you see how the content, audience, products, team, and money work together.
The full research prompt and the Nate Herk example so you can see what this looks like when it's finished.
A final section that helps you use the ideas without copying the creator.
How to use it
The order's important, so use it as it's written.
Pick the creator you want to study, then write down the one thing you're trying to figure out about their business.
Start with the creator's own website, videos, interviews, product pages, and public accounts. Don't start with somebody else's view of them.
Go through the ten parts in order. As you find something, mark whether it's a fact, something they reported, your own best read, an idea you could use, or something you still don't know.
Be careful with prices and money. A public price doesn't tell you what every customer paid, and it definitely doesn't tell you what the business made.
At the end, write down what could work for you, what would need to change, and what you shouldn't copy.
What it looks at
Ten parts of the business.
1
How they got started
What did they do first that made people pay attention?
2
How people find them
Where does the audience first come across their work?
3
What they make
What do they publish, where do they publish it, and what do they ask people to do next?
4
What people can use
Do they offer a guide, template, checklist, tool, course, or something else people can use?
5
Where the audience goes next
Can people join an email list, a group, a membership, or another place the creator can reach them again?
6
What they sell
What can somebody buy, how much does it cost, and what does it lead to next?
7
How the money might work
Which prices and sales numbers are public, which were only reported, and what still can't be known?
8
Who and what runs the business
What does the creator still do, what does the team handle, and which tools are actually being used?
9
How one part helps the next
How do the content, audience, products, feedback, and results keep helping each other?
10
What you can use yourself
Which ideas fit your situation, which need to change, and which should be left alone?
How the labels work
I needed a simple way to keep facts and guesses apart.
FACT
I found a solid source that backs it up.
REPORTED
The creator or company said it, but I couldn't verify it on my own.
INFERENCE
This is what the evidence seems to show, but they didn't say it directly.
ADAPTATION
This is an idea you may be able to change and use in your own work.
UNKNOWN
I couldn't find enough proof to say. So I don't.
REAL EXAMPLE / NATE HERK
Here's what it looked like when I used it.
I used the framework to break down Nate Herk's public business. I followed the path from his videos and free resources into AI Automation Society, AIS+, the Certified AI Consultant program, and the wider Uppit AI business. I wasn't trying to write his life story. I wanted to understand what each part did and how it connected to the next one.
Who people knowNate Herk
Where people find himYouTube and public content
What people can useTemplates, workflows, guides, and courses
Where people gatherAI Automation Society
Paid helpAIS+
Professional programCertified AI Consultant program
This case study only uses public information available when I did the research. It shows what I found, what Nate or his company reported, what I inferred, and what I still couldn't know.
# Creator Business Clone Deep Research Prompt
Version: 1.0
Resource: RS-001 Creator Business Clone
## Use this prompt
Replace `[CREATOR]` with the creator or creator-led business you want to study. Replace `[TARGET PERSON/BUSINESS]` with the person or business that will use the adaptation section.
The goal isn't to write a biography. It's to understand how useful work becomes attention, how attention becomes an owned audience, how that audience is monetized, and how customer or member activity feeds the next round of work.
## Prompt
Deep Research Task: Reverse Engineer the [CREATOR] Business and Personal Brand Ecosystem
Research [CREATOR], their companies, personal brand, content channels, communities, products, services, courses, books, software, partnerships, sponsors, affiliates, events, team, tools, past businesses, and publicly documented business results.
The goal isn't to write a biography. The goal is to reverse engineer the whole economic and attention system: how useful work becomes attention, how attention becomes an owned audience, how the audience is monetized, how customer or member activity feeds new products and content, and how the system has changed over time.
### Evidence rules
- Label direct evidence `FACT`.
- Label creator or company claims that aren't independently audited `REPORTED`.
- Label strategic interpretations `INFERENCE`.
- Label transferable recommendations `ADAPTATION`.
- Label unsupported or undiscoverable information `UNKNOWN`.
- Never invent conversion rates, traffic percentages, internal software, margins, sponsorship rates, member pricing mix, or revenue.
- You may show current sticker price multiplied by visible members as a scenario. Never call that actual revenue.
- Give every material number a source, date observed, and confidence level.
- Separate current evidence from historical evidence and estimates.
Use primary sources whenever possible. Start with the creator's current website, legal pages, checkout pages, and product pages. Then use platform-native pages, official partner pages, public interviews, reputable reporting, third-party analytics, and community discussion in that order. Treat pure inference as the weakest evidence.
### Research questions
Produce:
1. One-sentence business thesis
2. Month and year timeline
3. Brand and entity architecture
4. Attention engine by channel
5. Content strategy and performance patterns
6. Artifact engine
7. Audience capture and activation model
8. Actual funnel journeys
9. Full product ladder with current and historical pricing
10. Revenue tree separated into confirmed, reported, scenario, third-party estimate, and unknown
11. Team and founder delegation map
12. Tool stack classified as uses, teaches, sponsors, affiliate, partner, historical, or unknown
13. Content-to-product flywheel
14. Moats and network effects
15. Strategic evolution, including what the creator stopped doing
16. Risks and vulnerabilities
17. Fifteen structurally reusable principles
18. Five things another creator shouldn't copy
19. A 1-year, 3-year, and 5-year adaptation for [TARGET PERSON/BUSINESS]
20. Open questions and unresolved evidence gaps
### Required analysis
#### Origin and wedge
Find the smallest useful thing the creator became known for first. Separate the original wedge from the current brand.
#### Attention engine
For each channel, record audience size, cadence, format, discovery type, evergreen or trend role, destination, and strategic importance. Don't assign channel percentages unless public analytics support them.
#### Content engine
Separate topic choice, packaging, structure, voice, visual language, proof timing, CTA placement, resource placement, sponsor placement, series, repurposing, cadence, winners, and underperformers.
#### Artifact engine
Find the useful object that comes out of the content: a template, workflow, repository, checklist, dataset, prompt, calculator, scorecard, course, report, playbook, or tool. Record what created demand for it, where it lives, what the user exchanges to get it, and what happens next.
#### Audience and network
Map where the relationship moves after public discovery. Record entry friction, free value, activation, retention, peer value, moderation, and whether the audience surface has become a discovery channel of its own.
#### Value ladder and conversion
List every free and paid offer in order of commitment. For each one, record price, timing, buyer, promised outcome, delivery, next logical offer, friction, and evidence of scale. Follow real public journeys instead of inferring a funnel from navigation.
#### Revenue and economics
Separate visible current prices and scale, creator-reported results, mathematically implied scenarios, third-party estimates, and unknown revenue. Explain why a scenario may differ from realized revenue because of discounts, grandfathered plans, annual pricing, free accounts, refunds, churn, comps, or timing.
#### Operating system
Separate what the founder still owns from what has been delegated. Classify tools as internal use, teaching, sponsorship, affiliate relationship, official partner, historical use, or unknown. A sponsored tool isn't automatically the internal operating system.
#### Flywheel and moats
Map how real problems, experiments, results, content, artifacts, owned audience, questions, outcomes, new offers, proof, and authority feed each other. Score trust, owned audience, network effects, archive, IP, partner relationships, proof, team leverage, and switching cost from 0 to 5 when evidence supports a score.
#### Ethical adaptation
For every strategy worth considering, ask:
1. What problem does this solve for the original creator?
2. Does [TARGET PERSON/BUSINESS] have the same problem?
3. What existing asset could do the same job?
4. What's the smallest useful version?
5. What should stay off limits?
Separate what maps directly, what maps after modification, what doesn't map, what should wait, and what should never be copied.
### Finish with these compact artifacts
A. Ecosystem diagram
B. Revenue tree
C. Audience funnel
D. Compact timeline
E. Tool stack
F. Business machine scorecard
G. Fifteen lessons
H. Adaptation map
I. Five things not to copy
Before you finish, run an integrity check:
- Every material number has a source and date.
- Current and historical pricing are separate.
- Self-reported revenue is `REPORTED`.
- Scenario math isn't described as actual revenue.
- Third-party estimates are labeled.
- Sponsored tools aren't treated as internal tools without evidence.
- Unknown systems remain `UNKNOWN`.
- Audience percentages and conversion rates aren't invented.
- Strategy is separate from evidence.
- Adaptation is separate from imitation.
When the public evidence runs out, say `UNKNOWN`. That is a useful result.
## Practical usage note
Use this prompt in your own research. Adapt it to fit the subject and the question. Please don't resell, repackage, or present the original prompt as your own.
That's the full prompt. Copy it here or download the file above.
If you want more
Here's what I'm working on now.
The writing page has the longer pieces. The Climb is the weekly version of whatever I'm trying to figure out.