Operating by John Brewton: Oper(AI)ting by John Brewton

Build a Company of Curators

John Brewton — Tue, 06 Oct 2026 10:42:08 GMT

Start Here

Executives and leaders have an AI Slop problem, and they’re not happy about it.

If they receive one more 40-page report that is nothing more than an AI dump about an issue they’re dealing with internally, manufactured from half-clean data and with zero examination, clean summarization, or actionable direction by the deliverer of said file, there are going to be issues…

A 40-page report can contain useful information and most likely still be a poor piece of work to put in front of an executive. The numbers might be right. The research might be relevant. But the person receiving it still has to identify the important part and decide what to do with it.

Add several reports like that to a normal working week, and the time saved by the people producing them becomes work for the people reading them.

AI makes this easy to do. Give a model enough company data, and it can produce a document that looks finished long before anyone has decided what belongs in it.

Companies need to spend some of the time they save generating work on deciding what is worth sending.

The cost of an unfinished report

There is evidence of the cost of skipping that step. In a September 2025 survey by BetterUp Labs and Stanford’s Social Media Lab, 40% of 1,150 U.S. desk workers said they had received low-quality AI-generated work in the previous month.

Resolving an incident took roughly 2 hours on average, according to respondents. The researchers call this “workslop.”

Long documents deserve particular attention. In Zapier’s survey of 1,100 enterprise AI users, 52% identified long-form reporting among the tasks requiring the most cleanup or rework. Both studies rely on people’s reports of their experience, rather than an audit of company productivity. They still describe a recognizable problem: work arrives looking more complete than it is.

The version that concerns me goes beyond bad information. A report can have plenty of substance and very little judgment about what the reader needs. The author has gathered the material and passed the selection work upstairs.

Christian Catalini’s work on the economics of AI explains why verification becomes more valuable as execution gets cheaper. Companies need to know whether they can trust what their systems produce. Curation adds another obligation: deciding which of those trustworthy findings deserve attention in this particular decision.

An executive needs enough detail to make a sound decision, with the supporting work available when needed. That requires someone to understand the material well enough to organize it around the decision.

The report remains unfinished until that last bit of work is complete.

Use the time AI gives back

The time to finish it is part of what AI gives us back. In a 2023 randomized experiment, Shakked Noy and Whitney Zhang assigned writing tasks to 453 professionals. Access to ChatGPT reduced completion time by 40% and increased assessed quality by 18%. These were bounded writing assignments, so the figures shouldn’t be treated as a forecast for every company’s work. They show that faster production and better work can go hand in hand.

We need to make a deliberate choice about where the savings go.

If a first draft that used to take an afternoon now takes 30 minutes, use the remaining time to examine the recommendation. Check the source behind the number that drives it. Consider the alternative the draft dismissed too quickly. Then edit the document for the person who has to act on it.

I think of this as the final 20%: the work between a plausible draft and something you are willing to put your name on. The percentage is shorthand. For a difficult decision, that stage may take most of the time.

AI can help throughout it. It can compare alternative structures, identify repetition, and challenge whether a conclusion follows from the supplied evidence. We should use those capabilities to improve the final work. A faster first draft gives us room to raise the standard.

If you like this piece, you’ll love this one!

The 5 Curation Skills

Taste is the ability to recognize that standard in a particular context and explain why one version serves the reader better than another. In “The Overhang,” Ethan Mollick argues that, as production becomes easier, selecting what to keep and what to discard becomes a more valuable contribution. That’s a useful way to think about taste inside a company.

A finance team and a CEO may need different documents from the same analysis. The finance team needs enough detail to investigate. The CEO may need a recommendation about pricing and the evidence that would change it. Knowing how to serve each reader is a skill people can practice.

I would teach it through 5 curation skills.

1. Frame the decision

Before generating the report, finish this sentence: “After reading this, this person needs to decide _____.”

“An update on margins” leaves the assignment open. “Decide whether to change pricing on this product line before the next quoting cycle” gives the work a purpose. The reader’s responsibilities and the deadline determine what belongs in the brief.

For an informational update, name what the reader needs to understand and what would trigger action. There should be a reason to send it even when no immediate decision is required.

2. Judge the evidence

A confident paragraph and a well-formatted chart can make a weak conclusion look settled. Trace the claims that drive the recommendation back to their sources. Check the period covered and whether the data actually answers the question.

In a margin review, an observed decline is a fact. The explanation for that decline may still be an inference. A proposed price increase is a recommendation. Keep those distinctions visible, especially when the explanation is uncertain.

AI can help assemble a claim-and-source table. The author still needs to inspect the evidence behind the consequential claims. A second model agreeing with the first is insufficient on its own.

3. Rank what matters

Give the most consequential finding the most attention. A movement in a small product line may be interesting. A smaller movement in the company’s largest account may matter much more.

Ask AI to rank the findings against the stated decision and explain its ranking. Then challenge the order using what you know about the business. Financial size matters, but so can urgency or an obligation the model doesn’t understand.

The first page should make your priorities apparent. Equal space for every finding leaves the reader to establish the priorities themselves.

4. Connect the findings

Explain how the evidence leads to the proposed action.

In a hypothetical margin review, higher freight costs and lower margins might appear in separate sections. A useful analysis establishes whether freight explains the decline, how much remains unexplained, and whether changing prices would address it. Putting the observations next to each other doesn’t establish the connection.

Use AI to test the reasoning: have it identify the strongest supported objection and the missing evidence that could change the recommendation. Resolve those gaps or make them explicit in the brief.

5. Cut with care

Remove anything that doesn’t help the reader understand the decision, assess the evidence, or act. Move useful reference material into an appendix. Delete repeated background and findings included only because they were easy to generate.

Keep the uncomfortable fact that weakens your recommendation. Keep the uncertainty that could change the outcome. Curation becomes misleading when the editor removes the material a decision-maker most needs to see.

For each proposed cut, identify what the reader would lose. Sometimes the right answer is to keep the paragraph. A shorter document earns its length through better selection, not smaller type.

Use these 5 skills to select what the reader needs, including the evidence that complicates your recommendation.

Use AI to Help With the Edit

Employees can use AI for a separate editorial pass after the research and drafting. Give it the audience, the decision, and the source material. An instruction like this is much more useful than asking it to “make this better”:

Review this draft for the person who must decide [decision] by [date]. Identify the findings that could change that decision.
Show which consequential claims lack support in the supplied sources. Propose what to keep, move to the appendix, or cut, and explain each choice.
Preserve conflicting evidence and uncertainty that could affect the recommendation. Suggest a clearer order. Give me an edit plan before rewriting.

Review the proposed changes before accepting them. After the rewrite, compare the final brief with the source material and check what disappeared. The editing process can introduce omissions as easily as the drafting process can introduce errors.

Build curation into the job

This work needs a place in the job. A Microsoft Research and Carnegie Mellon study of 319 knowledge workers found that greater confidence in generative AI was associated with less reported critical thinking. It was a survey, not proof of skill loss. For a manager, it is a reason to make evaluation an explicit responsibility and give people the means to do it.

Leaders have to make the standard visible.

For a recurring executive report, agree on what “finished” means before the next version is commissioned. A useful default is a short opening brief containing the decision and recommendation, followed by the evidence needed to assess it. Material uncertainty belongs beside the recommendation. Supporting analysis should be easy to find. The document should name the person responsible for its conclusions.

Then show people examples. Keep a small library of strong briefs with annotations explaining the editorial choices. Include a before-and-after version so employees can see what was removed and why.

There is a useful research tradition behind this approach. Joanna Tai and colleagues’ work on evaluative judgment describes how learners can develop their ability to assess quality through examples and feedback. Applying that idea at work, I’d give employees 2 versions of the same brief and ask them to choose the stronger one, explaining their reasoning before a manager shares an assessment. That makes the company’s standards open to discussion and practice.

Protect some of the time saved for that work. If every faster draft earns the employee another assignment immediately, review will compete with a growing queue. An 8-month Berkeley Haas study at a 200-person technology company observed employees expanding their tasks and working longer hours as they adopted AI. That’s evidence from 1 company, but it is a useful warning against assuming that saved time will automatically translate into better judgment.

Return unfinished analysis with specific feedback and have the author revise it. Quietly rewriting everyone’s reports teaches dependence on the manager. Recognize useful selection, including the decision to stop producing a report nobody uses.

Managers should model the same discipline in what they send downward. Give a team a clear assignment with relevant context. Explain what needs their judgment. Forwarding an unreviewed AI analysis and asking everyone to make sense of it creates the same burden in the other direction.

Measure the result across the handoff. For 1 recurring report, record production time, review and correction time, and the recipient’s time spent clarifying it. Ask whether the brief supported the decision it was commissioned for. Faster drafting is a partial result until you know what happened to the people downstream.

Building a company of curators means developing this judgment throughout the organization. Start with the next recurring report. Have its author apply the 5 skills, use AI to challenge the edit, and get specific feedback from the recipient. Keep the changes that made it easier to use, and carry that standard into the next version.

- j -

Subscribe now

John Brewton documents the history and future of operating companies at Operating by John Brewton. He is a graduate of Harvard University and began his career as a PhD student in economics at the University of Chicago. Since selling his family’s B2B industrial distribution company in 2021, he has been helping business owners, founders, and investors optimize their operations.

How To Break Through The AI Adoption Wall

John Brewton — Mon, 28 Sep 2026 12:50:01 GMT

Give the right people a reason to buy.

People can enjoy your posts for months and still have no idea what they can hire you to do. That gap between attention and a clear offer is what we’re tackling in Master LinkedIn’s New Feed and Build an Offer People Want.

On September 29, I’m joining Alicia Teltz and Wessal Khader for a free, 90-minute masterclass covering how to:

Get your content in front of relevant buyers as LinkedIn’s feed changes.
Use customer insights to identify problems your audience wants solved.
Turn those insights into an offer people understand and want to pay for.

If you’re putting time into LinkedIn and struggling to connect that effort to something you can sell, join us.

Tuesday, September 29 · 1pm Eastern / 6pm UK · Free · Online

Start Here

This year, I’ve trained hundreds of employees inside companies to use AI and become AI-fluent. I’ve also worked with executives and teams to build their personal capabilities and the systems that support them.

The same problems keep showing up. Someone has access but cannot find a use for it. Someone else gets a poor answer and gives up. A capable employee quietly wonders whether getting better at this means helping the company eliminate their job.

Adoption is a management problem before it is a technology problem.

That is the argument behind The Adoption Wall. Its 10 hurdles fall into 3 groups: fear, habit, and context. Each needs a different response. Another demonstration of what the software can do will only get you so far.

I start with fear because it changes how people interpret everything that follows. An employee who thinks they are training their replacement has a sensible reason to hold back. Asking them to document their knowledge and automate their work without addressing that concern makes the request harder to trust.

My advice to leaders is to state what this program is for and how the company will use the capacity it creates. If you can commit that the program will not drive job cuts, say so early and keep that commitment. Name the bigger work people will take on. Never borrow trust with a promise you cannot keep.

Then there is the fear of looking incompetent.

Experienced people have spent years becoming the person others ask for help. A blank chat window can put them back in the position of a beginner, this time in front of their colleagues.

I make room for private questions and encourage leaders to show their own failed attempts. People need to see a poor first answer get corrected. That gives them permission to learn without treating every attempt as a public assessment of their ability.

Subscribe now

The confident expert presents a different challenge. They believe they are faster alone, and on some tasks they are. I use work they have already completed as the test. We have a known result and someone qualified to judge it. Then we compare the assisted approach, including the time needed to check and correct the output. The expert’s judgment sets the standard. A useful result earns another attempt.

Once people are willing to try, the next obstacle is how they work and improve.

One of the most common mistakes I see is asking for a substantial piece of work with almost no context. The request might be “write a proposal,” followed by disappointment that the proposal sounds like every other proposal.

I teach people to brief AI as they would a new employee. Explain the customer’s situation and the decision the document needs to support. Supply the constraints and an example of acceptable work. State what information is missing.

For people who struggle to write that brief, I suggest dictating it. Talk through the situation as if you were explaining it to a colleague. Then ask AI to organize the brief and identify gaps before starting the work.

Finding the right task is another difficulty. “Use AI in your job” leaves a busy person with the entire problem of deciding where to begin.

I ask them to log repeated work for 1 week. Look for information they keep reformatting, notes they keep organizing, or routine drafts they keep rebuilding. Start with a clerical task whose result they can check. The first useful application should be easy to recognize in their working day.

That also helps address the objection that there is no time to learn.

My approach is to bring real work into the session. Take something already due and build it together. A manager might prepare a meeting brief. A salesperson might turn approved notes into a follow-up draft. The learning happens while an existing piece of work gets done.

Managers also need to protect time for practice. If every hour remains committed to existing demands, people will return to the method they already know under pressure.

Then we have to make it happen again.

Old habits return when the deadline arrives, and the new method takes effort to reconstruct. I focus on replacing 1 recurring habit and saving the method as a reusable skill: the instructions, required inputs, expected output, and checks. Give it a clear name and keep it where the work happens.

The test comes the following week, when the person uses it without me.

This is where personal capability begins to become a system. Someone can explain the job clearly, judge the answer, and repeat the process. Their useful experiment now has a method another person can follow.

The final group of barriers concerns the context around the work.

Much of a company’s operating knowledge lives in people’s heads. An experienced employee knows which exceptions matter and when the usual process will fail. None of that appears in the request they give AI.

I use AI to interview the person about how they do the job. Have it ask 1 question at a time, work through an example, and probe the exceptions. Then turn the answers into a draft process for the employee to correct. Their review matters because a plausible process can still be wrong.

There is also the uncertainty about what is allowed.

Employees need usable answers about which tools they can use and what information they can share. I recommend a 1-page document covering approved models, permitted data, connector permissions, and where to take an uncertain case. It should point to the fuller policies where necessary.

Make the rules easy to find when someone needs them.

At the other end of the spectrum are the power users racing ahead alone. They build useful things, but the team cannot maintain a process it cannot find or understand.

I want a named owner, a shared library, and a clear route from experiment to production. Record what each workflow does and who checks its output. Before anyone depends on it, establish who maintains it and what happens when it fails.

A company can have several impressive AI users and still have no dependable way to share their work.

This is why I work on executive capability alongside employee training. Leaders need enough direct experience to judge whether an application is useful and to understand the work required to make it dependable. Delegating every experiment leaves them poorly placed to decide what deserves investment.

An executive’s own recurring task is a useful starting point. Build the method, use it again, and notice where it breaks. That experience makes the next conversation about team adoption more concrete.

For a team starting this week, I would choose 1 repeated task and 1 person to own the trial. Record the current time and quality standard. Build the assisted method using the approved information, then measure the time spent on checking and correction.

Save what works. Run it again the following week. Ask someone else to follow the instructions and record where they get stuck.

Those observations tell you what to fix next: the brief, the process, the rules, or the training. They also give the next session a task.

Put that review on the calendar before the first session ends.

Concluding Thought

After training hundreds of employees this year, the lesson I keep coming back to is that AI adoption depends on what happens after the training. People need to feel safe enough to try, understand where the tools fit their work, and have a method they can use again. Leaders are responsible for creating those conditions and helping useful individual experiments become something the whole team can rely on.

Start with 1 recurring task this week. Set the quality standard, build the assisted method, and count the time spent checking and correcting it. Save the instructions and ask someone else to follow them the next week. When another person can produce acceptable work with less effort, you have evidence of progress. That is what I want companies to leave training with: the capability to keep improving the work themselves.

The models and technology will continue to improve exponentially. That’s why the training should prioritize driving meaningful changes within the organization, enabling the team to move forward and evolve into a company that fosters continuous learning, improvement, and adaptation.

Hope this helped you today. If you have a company that is looking to make these changes, I’d love to talk with you. If you are a consultant or coach who is in the midst of doing this work with your clients, be sure to drop me a message as well.

- j -

Give the right people a reason to buy.

On September 29, I’m joining Alicia Teltz and Wessal Khader for a free, 90-minute masterclass covering how to:

Get your content in front of relevant buyers as LinkedIn’s feed changes.
Use customer insights to identify problems your audience wants solved.
Turn those insights into an offer people understand and want to pay for.

If you’re putting time into LinkedIn and struggling to connect that effort to something you can sell, join us.

Tuesday, September 29 · 1pm Eastern / 6pm UK · Free · Online

Take Your Seat

Apparently AI is going to kill us all. Finally, someone figured out how to write a hook.

John Brewton — Thu, 24 Sep 2026 10:37:25 GMT

Give the right people a reason to buy.

On September 29, I’m joining Alicia Teltz and Wessal Khader for a free, 90-minute masterclass covering how to:

Get your content in front of relevant buyers as LinkedIn’s feed changes.
Use customer insights to identify problems your audience wants solved.
Turn those insights into an offer people understand and want to pay for.

If you’re putting time into LinkedIn and struggling to connect that effort to something you can sell, join us.

Tuesday, September 29 · 1pm Eastern / 6pm UK · Free · Online

Start Here

Dario Amodei’s Anthropic has built its Responsible Scaling Policy around catastrophic AI risks, including biological threats and systems acting beyond human control.

In an interview published September 3, Sam Altman called the next generation of models “sobering for everybody.” He later warned about 2 outcomes: humanity losing control of AI, and AI concentrating power in a single person, company or country.

Elon Musk has been sounding the alarm for years. In 2023, he signed an open letter calling for a pause in training the most advanced systems.

Getting Amodei, Altman and Musk to broadly agree on anything may itself qualify as an extinction-level event.

People often need a reason to care before they need a reason to buy…

The companies warning us about AI also have enormous commercial interests in the rules governing it. Those rules help determine who can compete and what competing will cost.

A 2023 paper coauthored by OpenAI researchers proposed registration, reporting, safety standards and enforcement for frontier-model developers. Anthropic has advocated a “regulatory ladder”, with testing and government oversight increasing as the risks grow.

Those may be sensible requirements. Depending on their design, they may also be considerably easier for an established lab to meet than for a smaller challenger.

The thresholds matter. A rule aimed at developers of the most capable models does not automatically apply to everyone building an AI app. But compliance costs can still shape who gets to challenge the leaders.

Brookings has warned that licensing can reinforce incumbents by raising the cost of entry. Separately, the Federal Trade Commission has identified how control over essential AI inputs can entrench established firms.

That is a reason to examine the incentives. It is not evidence that these executives are manufacturing fears they do not believe.

The risks can be real, the concerns sincere, and the regulations justified. The same regulations can still make life harder for future competitors.

A warning can serve the public interest and the company delivering it.

Fear can draw attention to a real problem. Attention can influence which solutions get heard. And the proposed solution can have commercial consequences of its own.

Most founders face a considerably less apocalyptic version of the first challenge: getting somebody to care.

Subscribe now

Here are 10 lessons worth taking from these recent shenanigans

1. Start with the problem

Nobody wakes up hoping to learn about your proprietary methodology.

They wake up with problems.

The AI warnings begin with consequences people can understand: a cyberattack, a biological threat, a system humans cannot control. The technical explanation comes after the reason to listen.

Your customer deserves the same order.

“We offer a 6-week advisory engagement” describes something you sell.

“Your sales team cannot explain why someone should choose you” describes something a buyer might need to fix.

Start where the customer is already uncomfortable.

2. Make the stakes obvious

There is a reason nobody leads with:

AI could marginally inconvenience humanity.

A problem earns attention when its consequences become clear. For your customer, those consequences might be lost revenue, shrinking margin, wasted labor, customer churn, or another year spent solving the wrong problem.

“Our quoting process is slow” is easy to postpone. “Customers are choosing someone else while we’re still preparing the quote” gives the delay a cost.

Find that cost. Use numbers the customer can verify.

3. Make it about them

22 years of experience.
6 certifications.
A Harvard degree.
7 modules.
43 templates.

Wonderful. But nobody cares.

Your buyer is still trying to work out what changes for them.

Credentials can establish that you are qualified to help. They cannot do the job of explaining why the help matters.

“You’ll leave with a price you can defend in a sales conversation” gives the buyer something to picture. “You’ll get access to 7 modules” gives them something else to finish.

The customer should recognize their own situation before you ask them to admire yours.

4. Create tension

“AI is getting better” contains information.
“AI may become capable faster than we learn to control it” gives you a reason to keep reading.

The tension sits in the gap between progress and preparedness. In your business, it might sit between the effort a customer is making and the result they keep failing to get.

Consider a founder buying more traffic for an offer nobody understands. The useful observation is simple: more visitors will encounter the same confusion.

Now the founder has a reason to reconsider the offer before buying the next ad.

Good tension exposes something the customer needs to resolve. It earns the next sentence.

5. Get specific

Drama earns attention. Specificity gives people something to evaluate.

“We’ll transform your business” asks the buyer to supply the meaning.

Compare that with:

In 30 days, you’ll have a tested offer, a price, positioning, sales copy, and a live way for customers to buy.

Now the buyer can picture the work and judge whether it is worth doing.

A promise this specific also gives the customer a way to hold you to it. Define what “tested” means. Make sure the timeline and deliverables are ones you can stand behind.

6. Give people a way forward

This is where the psychology gets useful.

A 2015 meta-analysis covering 127 papers, 248 samples and 27,372 participants found that fear appeals had a positive average effect on attitudes, intentions and behavior. They worked better when the message also explained that the recommended action could work, or that the person could carry it out.
A 2022 meta-analysis covering 19,736 participants found stronger effects on behavioral outcomes when messages included information that the recommended response would be effective. Most of its messages concerned health, so applying the finding to sales requires judgment.

My practical reading: if you make a problem vivid, make the response credible too.

“Your sales are slipping” creates concern. A clear diagnosis, a workable first step, and evidence from a comparable situation give the customer something to do with it.

Show them the door. Explain why it leads somewhere.

7. Make the path simple

Once someone cares, don’t hand them a maze.

For an offer-building engagement, the path might be:

Find it. Name it. Validate it. Price it. Publish it. Sell it.

That is easier to follow than “participate in our iterative commercialization methodology.”

The work behind each step may be complicated. The customer still needs to understand where they start, what happens next, and what they will have when they finish.

Describe the path in words they can use without you in the room.

8. Create real urgency

If a problem costs $30,000 every month, another month of delay has a price. If a customer keeps losing bids because quotes arrive late, waiting means accepting more of the same.

That is enough to discuss urgency without installing a countdown timer.

Use the customer’s actual costs and deadlines. A season ending, a contract renewing, or another month of avoidable losses can give them a sound reason to act.

Invented deadlines teach buyers to doubt the next deadline too.

9. Make the idea repeatable

“AI might kill us.”

4 words. Try forgetting them.

Now compare that with:

“We leverage cross-functional technology-enabled capabilities to unlock enterprise transformation.”

Try remembering it.

Your customer should be able to explain what you do to someone else without reopening your website. That matters when the person you persuaded has to persuade a partner, a boss, or a budget holder.

Give them a sentence they can carry into that conversation.

10. Earn attention without betraying trust

Once you discover that threats attract attention, the temptation is obvious: make everything bigger, scarier, more urgent.

Turn every inconvenience into an emergency.

Eventually, the customer learns to discount you.

Describe a real problem vividly. Keep the scale of the claim proportionate to the evidence. Make the cost of waiting clear, then give the buyer room to decide.

Great marketing makes genuine importance difficult to overlook.

Your next claim will arrive carrying the reputation of your last one.

Give them a reason to use the pen

Consider the familiar sales test: “Sell me this pen.”

Most founders would start with the ink and the craftsmanship. Then the materials, the founder’s résumé, and the proprietary pen framework.

6 pen modules. A private pen community.

Meanwhile, the buyer still needs a reason to write something down.

That is the job these 10 lessons help you do. Identify a problem the customer recognizes, make its consequences clear, and offer a credible way forward. Make the message simple enough to remember and honest enough to trust.

Ideally, you can manage this without threatening the extinction of the human species.

Attention gives you an opening. You still have to earn the sale.

Before you publish your next offer, underline the first sentence that gives the customer a reason to care. Check that the consequence is real and the promised response is one you can deliver.

If you have to read halfway down the page to find that sentence, move it to the top.

Give the right people a reason to buy.

On September 29, I’m joining Alicia Teltz and Wessal Khader for a free, 90-minute masterclass covering how to:

Get your content in front of relevant buyers as LinkedIn’s feed changes.
Use customer insights to identify problems your audience wants solved.
Turn those insights into an offer people understand and want to pay for.

If you’re putting time into LinkedIn and struggling to connect that effort to something you can sell, join us.

Tuesday, September 29 · 1pm Eastern / 6pm UK · Free · Online

Pick Your Business Model Like Your Life Depends On It.

John Brewton — Fri, 18 Sep 2026 10:24:55 GMT

Hello Operators!

Most founders do not really choose a business model. They copy one.

They watch someone build a course, so they start outlining a course. They see a consultant with five retainers, so they decide they need retainers. Someone launches a cohort. Someone else sells a $10,000 advisory package. Someone licenses their IP.

What was visible becomes what feels possible.

The problem is that a business model is not a trend, a personality type, or an identity. It is the delivery system for your expertise. And the right delivery system has to fit two things at the same time: the shape of what you know and the shape of the life you actually want to live.

Today’s free Oper(AI)te drop is built to sort those two things.

FREE DROP 05: The Model Sorter

Download Today's Drop

This is not really a workbook you read. Upload the PDF into Claude or ChatGPT. Paste one sentence describing what you do. The workbook takes over from there.

You give it one capability and answer a short interview. It scores six business models against your constraints, shows you the grid before it sorts, lets you correct anything you disagree with, and then produces a one-page PDF with the result.

The workbook estimates the conversation at about seven minutes. Go get to work.

Subscribe now

Your Expertise Is Only Half the Business Model

The first four Oper(AI)te drops were largely about evidence.

The Calendar Autopsy looked at what your work has actually required you to do.

The Receipt Finder turned an accomplishment into a claim you can defend.

The Compliment Decoder looked for the language other people already use to describe your value.

The Free Advice Audit looked for expertise you may be giving away because it feels too easy to you.

If you have been running them, you are beginning to accumulate something useful: a picture of what you are unusually good at and evidence that you have actually done it.

But knowing what you could sell does not tell you how you should sell it.

That is a different decision.

And founders routinely make it without considering the constraints that will eventually determine whether they love or hate the company they built.

The Workbook Output

The final output is deliberately small, clear, crisp and clean.

Your top two models appear with their scores out of four and the constraint that separated them. You get the one model the Sorter ruled out and why it made that choice.

Then you get something I find more useful than another list of advantages and disadvantages:

The characteristic failure mode of your winning model.

Productized service fails when scope creeps until every engagement becomes custom again, but the price never moves with it.
Retainer or advisory fails when the founder slowly becomes the client’s employee without the salary, equity, or exit.
Licensing fails when nobody wants to buy a method they have never watched the founder run, so the founder keeps running it.
A cohort can turn your calendar into a permanent launch treadmill.
A self-serve course can quietly turn the founder into a full-time marketer.
The diagnostic can get given away in an attempt to win an implementation that never closes.

And then the workbook ends with one clean, clear, crisp line:

Your model is X. The one you were about to build is Y.

Download Today's Drop

Where the Free Drops End and Oper(AI)te: The Hidden Offer Begins

The Model Sorter can help you answer one important question:

What shape should this business take?

It cannot build the rest of the business for you.

That is the distinction between these free drops and the Oper(AI)te: The Hidden Offer cohort Wessal Khader and I are building.

The drops isolate one problem and give you one useful output. The October program connects the evidence, expertise, model, market, pricing, offer, sales page, and selling system into one operating system.

Most experienced founders, operators, executives, consultants, and advisors do not have an expertise problem. They have a problem seeing, packaging, pricing, and selling what they already know.

That is our entire premise behind Oper(AI)te: The Hidden Offer.

Take Your Seat

The offer someone will pay you for is probably not something you need to invent from scratch. It is much more likely to be sitting inside the work you have already done, the results you have already produced, the problems people already bring you, and the expertise you have accumulated so gradually that you no longer recognize how unusual it is.

From October 1 through October 29, we are building that system live across six sessions. You build nine AI agents on our IP and keep them: the Everything File, Superpower Scan, Fit Test, Client Evidence Pull, Market Map, Intersection Engine, Price Evidence Sheet, Page Builder and Language Audit, and the Selling Agent with its Daily 15 rhythm.

The goal is to exit October with your offer ready to launch as this year ends and 2027 awaits!

Here’s what our cohort members will leave with:

Your evidence gathered.
Your expertise named.
Your model selected.
Your market researched.
Your price supported.
Your sales page published.
Your first buyer conversations underway.
And a 90-day operating rhythm for continuing the work after the cohort ends.

See everything inside Oper(AI)te: The Hidden Offer →

Doors close September 30.

The live cohort runs October 1–29.

I am keeping all 16 free Oper(AI)te drops together here and updating the library as each one goes live.

Because your ideal offer is probably not something you need to dream up. And the business model probably should not be something you copy.

Both should come from the evidence of the work you’ve already done, been paid for, and delivered immense value to your companies, teams, colleagues, and customers with already.

Take Your Seat

My Biggest Give Away Yet: The Oper(AI)te (Hidden Offer) Drop Library

John Brewton — Wed, 16 Sep 2026 11:17:31 GMT

Subscribe now

For the next two weeks, I am giving away one of the operating tools Wessal Khader and I have been building around Oper(AI)te: The Hidden Offer, every morning.

Each drop is a small AI-powered operating system built to take something you already have (your calendar, an old client result, emails people have sent you, your pricing, a competitor’s website) and turn it into a meaningful insight or asset that will help you to build a better offer and a better business to operate.

A piece of evidence.
A clearer understanding of what you are unusually good at.
A pricing insight.
A case study.
A list of real prospects.
A better way to explain what you sell.

My goal is to give you something valuable enough to change how you see the expertise you already own. I created this article to keep every drop in one place. I will update it each morning as the next one goes live.

Bookmark it. Save it. Send it to somebody who has spent a career becoming excellent at something and still has not figured out how to turn that expertise into an offer.

How to Use the Drops

1. Download the workbook.

2. Upload it into Claude or ChatGPT.

3. Tell the model to begin the workbook.

Then follow the conversation.

The workbook will ask the questions, analyze the information, and produce the output. Some drops work entirely from information you provide. Others use your files, connected tools, live research, or material you have already created.

Do not overthink them.

Run them on your own intuition and instinct.

The value is not in understanding what the workbook does. The value is in seeing what it discovers about you.

What These Drops Are Actually Building

There is a larger reason Wessal and I built these 16 workbooks.

Most experienced founders, operators, executives, consultants, and advisors do not have an expertise problem. They have a problem seeing, packaging, pricing, and selling what they already know.

That is our entire premise behind Oper(AI)te: The Hidden Offer.

Take Your Seat

These 16 drops are designed to help you start finding it.

Over the next two weeks, we are going to move through a very deliberate sequence:

Find the proof.
Identify the expertise.
Understand the buyer.
Check the market.
Establish the price.
Articulate the offer.
Build the sales assets.
Start selling.

But there is an important distinction.

The drops give you pieces. The course builds the system.

The free drops take one problem, one input, and one pass and give you something useful back. Inside Oper(AI)te, we connect those discoveries.

Instead of examining one quarter of your calendar, you build an Everything File that lets AI understand the body of work you have built across your career.
Instead of identifying a single piece of valuable expertise, the Superpower Scan and Fit Test help determine what you should actually build a business around and which business model fits it.
Instead of focusing on a single competitor or a single pricing question, the Market Map and Price Evidence Sheet examine what buyers are paying right now and what your evidence can support.
Instead of writing a better sentence, the Intersection Engine, Page Builder, and Language Audit turn the evidence into an offer and put it onto an actual sales page.
And instead of finding five prospects once, the Selling Agent and Daily 15 are built to help make selling part of how the business operates every day.

There are 9 AI agents in the complete system. You build with them live, and you keep them. By the end of the October program, the goal is not for you to simply understand more about offers.

The goal is for you to leave holding one.

Your evidence gathered. Your superpower named. Your business model selected. Your market understood. Your offer stated clearly. Your price supported by evidence. Your sales page live. Your selling system operating. Your first buyer conversations underway.

But you do not need to take my word for any of this yet.

Run the free tools first.

If they show you nothing useful, keep the workbooks. If they begin uncovering things you have spent years overlooking, pay attention.

Those discoveries are not random.

They are your evidence.

Eventually the question becomes:

What happens if you connect all of it?

For now, let’s start with the work.

AVAILABLE NOW

FREE DROP 1

The Calendar Autopsy

Your title tells people what your job was called. Your calendar shows what you were actually paid to do.

Your résumé is aspirational. Your LinkedIn profile is curated.

Your calendar is harder to fake.

Open 90 days of it and you will find a surprisingly honest record of what an organization actually depended on you to do.

The Calendar Autopsy analyzes the problems you repeatedly spent time solving, the capabilities that kept pulling you into rooms, the work consuming the largest share of your time, and which of those capabilities sat closest to revenue.

It then asks a much more interesting question:

Does the work your title says you do match the work your calendar says people actually needed you to do?

For many experienced operators, the answer is no.

That gap matters because the work people repeatedly pull you into is often much closer to your monetizable expertise than the generic description sitting at the top of your LinkedIn profile.

What this drop finds

The capabilities people repeatedly relied on you to provide and the evidence of what your real job actually became.

What it doesn’t know yet

Which of those capabilities a market would pay for, what form the offer should take, or which piece of your experience should become the business.

Download The Calendar Autopsy →

One note: this workbook works best when the model can access your calendar or when you provide the last 90 days of calendar data.

FREE DROP 2

The Receipt Finder

Turn “I helped them improve the business” into a claim a buyer can actually evaluate.

Experienced people are often surprisingly bad at describing what they have accomplished.

We say things like:

I helped grow the company.

I improved operations.

I helped the team scale.

I worked on the transformation.

All of those statements may be true. Almost none gives a buyer enough information to understand what actually happened.

The Receipt Finder starts with one accomplishment and forces you to establish four things:

What was true before you arrived?

What was true afterward?

Over what period did the change happen?

What role did you actually play in creating it?

Then it goes one step further. The workbook builds multiple versions of your claim based on what your evidence actually supports.

A conservative version. An accurate version. And the strongest version you can currently defend.

That distinction matters.

Because I do not think the answer to selling expertise is making enormous claims. I think the answer is becoming unusually precise about the value you have actually created.

You leave with language you can use in a buyer conversation and a clear list of the evidence you would need before making a stronger claim.

What this drop finds

The measurable result hiding underneath a vague accomplishment and the strongest version of that result your current evidence supports.

What it doesn’t know yet

Which of your dozens of career accomplishments matters most to the buyer you eventually decide to serve.

Download The Receipt Finder →

FREE DROP 3

The Compliment Decoder

Other people may already have written your positioning for you.

One of the hardest transitions in building a coaching, consulting, advisory, or expertise-based business is becoming comfortable selling yourself as the service.

We tend to believe we need a library of polished testimonials and professionally produced case studies before we have enough proof. But much of the proof already exists.

It is sitting inside old emails from clients, thank-you notes from colleagues, LinkedIn recommendations, performance reviews, referral introductions, Slack messages, texts, and sentences people have written for years about what you do unusually well.

The Compliment Decoder turns that unstructured feedback into evidence.

Give it 10–20 pieces of real feedback and it begins looking for repetition. Which capabilities keep appearing? What do people actually rely on you for? What language do they naturally use when describing the value you create?

And there is an important distinction buried inside that analysis:

What people admire about you is not always what they rely on you for.

The workbook separates those two things. Then it looks for something even more useful: the blind spot.

The thing other people consistently recognize in you that barely appears in the way you describe yourself.

That can be uncomfortable. It can also be incredibly useful.

Because suddenly your positioning is no longer based entirely on what you hope you are good at. It is grounded in what other people have been telling you for years.

What this drop finds

The recurring themes and exact language other people already use when they describe the value you create.

What it doesn’t know yet

Whether that valued capability maps to a real market, a specific buyer, a viable business model, or a problem somebody will pay to solve.

Download The Compliment Decoder →

FREE DROP 4

The Free Advice Audit

The thing you give away in 20 minutes may take somebody else six months to learn.

There are probably a handful of questions people ask you constantly.

“Can you take a quick look at this?”

“What would you do here?”

“Can I get your opinion on my pricing?”

“Who would you hire?”

“Can I pick your brain for 20 minutes?”

And because answering feels easy, you answer. Usually for free.

That ease is exactly the problem The Free Advice Audit is designed to expose.

Something that takes you 20 minutes may take the person asking you weeks or months to solve without you. That difference is not trivial.

It is the accumulated value of your experience, pattern recognition, mistakes, judgment, relationships, and expertise. And in some cases, it may be the beginning of an offer.

The workbook asks you to identify the five things people most frequently come to you for and then compares them across three dimensions:

Effort asymmetry: How long does solving this take you compared with them?

Consequence: What happens if they make the wrong decision?

Frequency: How often does this problem occur?

The workbook ranks all five. Then it identifies the strongest candidate and gives it a very specific label:

This is a product, not a favor.

But my favorite part is the final number.

The workbook looks at your own hourly value and estimates how much of this expertise you gave away during the last 12 months. It puts a dollar figure on the advice.

That number may sting a little.

I hope it does.

Because one of the easiest things in the world to undervalue is knowledge you acquired so slowly that you forgot how difficult it was to learn.

What this drop finds

The piece of expertise with the greatest gap between how easy it feels to you and how valuable it may be to somebody else.

What it doesn’t know yet

Whether anyone will buy it, which buyer needs it most, what form it should take, how it should be positioned, or what that buyer will pay.

Download The Free Advice Audit →

FREE DROP 5 · SEPTEMBER 17

The Model Sorter

Download Drop 5

Most people choose a business model by imitation. They see somebody selling a course, coaching program, retainer, advisory engagement, or productized service and decide that must be the model they need too.

The Model Sorter starts somewhere else: with the shape of your expertise and the shape of your life.

It looks at your available time, appetite for synchronous work, how dependent the value is on your personal judgment, whether the customer problem repeats, and other constraints. Then it compares different business models and tells you not only which models fit, but which one you should probably rule out.

FREE DROP 6 · SEPTEMBER 18

The Buyer’s Real Words

Download Drop 6

You describe the customer’s problem using your vocabulary. Your buyer describes it using theirs.

Those two languages often barely overlap.

The Buyer’s Real Words researches the places your customer talks naturally and pulls out the language they use to describe the problem, what they have already tried, and why those attempts failed.

You leave with language you can use in positioning and sales because it came from the market instead of your imagination.

Take Your Seat

COMING NEXT…

FREE DROP 7 · SEPTEMBER 19

The 90-Second Teardown

You have probably looked at competitor websites. You may never have actually read one.

Give this workbook a competitor URL and it breaks the page into the promise, buyer, proof, pricing logic, objection handling, and something most competitive analyses ignore:

What the competitor never talks about.

That silence may be where your differentiation is hiding.

FREE DROP 8 · SEPTEMBER 20

The Price Spread

Three people can appear to sell roughly the same work for $2,000, $15,000, and $80,000. The work is rarely 40 times different.

Something else is creating the spread.

The Price Spread researches real providers across different pricing tiers and begins identifying what the buyer is actually paying more for: risk transfer, access, proof, buyer type, outcome ownership, contract structure, category framing, or some combination of them.

The question changes from “How can they charge that?” to “What would have to become true for me to occupy that tier?”

FREE DROP 9 · SEPTEMBER 21

The Underpricing Calculator

Most experts price the work. Buyers value the outcome.

Those are two entirely different numbers.

The Underpricing Calculator puts your current price beside the financial consequence of the outcome you create and calculates the relationship between the two. Not to manufacture the biggest number possible, but to find the number your existing evidence can actually support.

FREE DROP 10 · SEPTEMBER 22

The Generic Detector

You cannot hear your own filler. Most of us have bios, About pages, and LinkedIn profiles full of phrases thousands of competitors could use word for word.

The Generic Detector applies a simple test:

Could a direct competitor say this exact sentence?

If the answer is yes, the sentence is carrying almost no information. The workbook flags the generic language, calculates how much of your writing is truly specific to you, and begins replacing the filler with evidence from your actual work.

FREE DROP 11 · SEPTEMBER 23

The One-Sentence Compressor

If it takes three paragraphs to explain what you sell, you probably do not have an offer yet. You have a description.

The One-Sentence Compressor takes your current explanation and creates several structurally different versions, then tests them for specificity, buyer recognition, clarity, and one criterion I especially like:

Could somebody else accurately explain what you do after hearing it once?

Because an offer that cannot travel without you is much harder to sell.

FREE DROP 12 · SEPTEMBER 24

The Proof Page

Most experienced operators already have work that could help close their next customer. Nobody can see it.

The Proof Page takes one past project and turns it into a clean, client-facing case study built around the situation, the constraint, what you did, the defensible result, and what that result demonstrates you can do again.

Unlike most of the drops, this one does not only give you an insight.

It gives you something you can send to a buyer that day.

FREE DROP 13 · SEPTEMBER 25

The Objection Pre-Mortem

You are going to hear many of the same objections repeatedly once you start selling an offer. You can wait for the next 20 conversations to discover them, or you can prepare now.

The Objection Pre-Mortem starts with your actual offer and price, identifies the objections most likely to appear, separates the surface objection from the underlying concern, and determines what evidence would answer it.

You leave with something even more useful than an objection list:

An inventory of the proof your business still does not have.

FREE DROP 14 · SEPTEMBER 26

The First Five

“Find me leads” is a terrible instruction.

“Who specifically appears to have this problem right now, and what is the evidence?” is much better.

The First Five starts with your offer and buyer and looks for five real companies or people with a reason the problem may be live now: a new hire, funding round, leadership change, job posting, launch, complaint, regulatory shift, or another observable trigger.

Five prospects. Five reasons why now.

FREE DROP 15 · SEPTEMBER 27

One Day of the Daily 15

This drop is not really another diagnostic. It is a protocol.

One post. One new conversation. One follow-up. Fifteen minutes.

You use the work produced by the previous drops to execute one complete day of a simple selling rhythm.

The interesting part is not whether you can do it once.

The interesting question is what happens when the business can help you do it every day.

FREE DROP 16 · SEPTEMBER 28

The Headline Stress Test

You can spend months developing an offer and four minutes writing the headline that introduces it.

The final numbered drop takes your offer and produces 12 different headline structures, then scores them for specificity, buyer recognition, clarity, and how much meaningful information appears in the first few words.

You finish with page-ready language.

But by then, the headline will almost be beside the point.

Because look at what you will have accumulated.

What You Will Hold After 16 Drops

You will have a much clearer description of the work people have actually relied on you to do, along with a defensible career result and a bank of language other people use to describe your value. You will also have evidence of expertise you may have been giving away for years without realizing how commercially valuable it could be.

You will have a business model that better matches how you actually want to work. You will understand more of your buyer’s language, your competitive environment, and what creates pricing power in your category.

You will have a better view of the relationship between your price and the value you create, more specific positioning, and a one-sentence offer. You will also have a finished case study, a clearer view of your likely objections, and an inventory of the evidence required to answer them.

You will have five real prospects with reasons to contact them now. You will have a simple selling rhythm and page-ready language to put around the offer.

Sixteen useful outputs, all built from your career, your evidence, your market, and your buyers.

And then we arrive at the most important part.

What you still will not have is anything that runs.

You will have performed the analysis and produced the individual pieces. But you will not yet have one AI business brain that holds that evidence together, updates it, and uses it across the rest of the business.

You will not have agents continually reading that evidence or a Market Map designed to be rerun as the category changes. You will not have a permanent Price Evidence Sheet connecting your market, proof, and pricing decisions.

You may have the pieces of an offer, but not yet the complete offer assembled from the intersection of your proof and actual market demand. You may have stronger copy, but not necessarily the sales page published and working as a conversion surface.

And you will not have a Selling Agent waking up with you every morning and turning all of that thinking into activity.

That difference is Oper(AI)te: The Hidden Offer.

The drops are designed to let you experience pieces of what Wessal and I have been building. The October cohort is where we assemble those pieces into the operating system.

From October 1–29, we build it live.

You will build the Everything File, the Superpower Scan, the Fit Test, the Client Evidence Pull, the Market Map, the Intersection Engine, the Price Evidence Sheet, the Page Builder and Language Audit, and the Selling Agent with its Daily 15 rhythm.

That is 9 AI agents built on our IP, yours to keep.

Across six live sessions, you will build against your actual career, business, expertise, market, and buyers rather than watching us demonstrate the process against a hypothetical company. The objective is for the work to leave the classroom and become something operating inside your business.

Your sales page gets published. Your first real buyer conversations get opened and coached, and you leave with a 90-day rhythm for continuing to operate the system after the cohort ends.

The finish line is not, “I learned how to build an offer.”

It is being able to say: I have an offer. I know who it is for and why they need it. I know what evidence supports it and what I can defend charging for it.

You can send somebody the page.

And you have a system helping you sell it.

If these free drops are already showing you things about your career and expertise you had not seen before, that is the best explanation I can give you for why we built the full program. The question is no longer whether the expertise exists.

The question is whether you want to turn it into something that operates.

See everything inside Oper(AI)te: The 30 Day Offer Launch →

Doors close September 30, and the live cohort runs October 1–29. Until then, keep running the drops and keep paying attention to what the evidence is telling you.

Drop 5 goes live tomorrow morning.

This page will keep getting better every day.

Harvard Taught Me How to Negotiate in Theory. $250 Million in Deals and 20 Years of Work Taught Me the Rest.

John Brewton — Mon, 14 Sep 2026 11:42:05 GMT

Oper(AI)te: The Hidden Offer

Oper(AI)te™: The Hidden Offer is a four-week live cohort that helps experienced founders and operators turn the expertise they’ve already built into one clear, market-validated offer with a defensible price. You leave with the offer named, priced, published on a live sales page, and supported by AI tools designed to help you start buyer conversations.

Take Your Seat

The Beginning

While I was an undergraduate at Harvard studying economics, one of the most useful things I studied had surprisingly little to do with economics. I had the opportunity to study negotiation through Harvard Law School and the intellectual tradition around the Harvard Negotiation Project, the work that produced ideas like Getting to Yes, BATNA, interests versus positions, objective criteria, and creating options for mutual gain.

At 20-something years old, these ideas felt elegant. Twenty years later, they feel expensive, because I got to spend much of the next two decades testing them with real companies, real contracts, real jobs, real inventory, real payroll, real margins, and hundreds of millions of dollars at stake.

I eventually became CEO of an industrial distribution business. Over 12 years, we grew the company from roughly $27.5 million to more than $75 million in annual revenue before selling it in 2021. Negotiation sat underneath almost everything we did.

I negotiated with companies like Johnson Controls, All-Clad Metalcrafters, and more automative, aerospace and defense OEMs than I can count. Over time, I negotiated and structured more than $250 million in enterprise contracts, including customer relationships that grew from roughly $750,000 to $13 million in annual revenue for the company. It was an extraordinary learning experience not only in negotiating deals, but also in understanding how offers for services and products are built and optimized.

These were rarely simple negotiations over a number on a purchase order. They involved engineering requirements, sourcing, inventory, quality, capacity, tooling, lead times, service levels, payment terms, logistics, forecasting, capital, risk, contractual commitments, and relationships that could last for years.

There was another important difference: I often had to operate the deal after I negotiated it. That may be the single biggest thing that changed how I think about negotiation. A salesperson can celebrate the signed agreement. An operator eventually discovers whether the agreement was any good.

Harvard Gave Me the Language

The Harvard approach to negotiation begins with several deceptively simple principles: separate the people from the problem, focus on interests rather than positions, create options for mutual gain, rely on objective criteria rather than arguing over whose number is right, and understand your BATNA—your best alternative if an agreement cannot be reached.

I still believe all of those things. What operating taught me was what they look like after they collide with reality.

Take the classic distinction between a position and an interest. A customer says, “We need the price reduced by 10 percent.” That is a position. The more important question is why.

Maybe procurement has a savings target. Maybe another supplier quoted less. Maybe the customer’s budget changed. Maybe somebody internally needs to demonstrate a concession before approving the contract. Maybe total price matters less than cash flow. Maybe the real issue isn’t price at all but inventory, lead time, quality, capacity, service, or uncertainty about whether you can deliver.

If you negotiate only the position, you have one variable: price. Understand the interest and you may suddenly have ten.

That distinction probably saved me millions of dollars during my career. It also led me toward one of the first additions I made to what I had learned academically: before you concede on one variable, create more variables.

A customer wants a lower price. Fine. What changes with it? Volume? Payment terms? Length of commitment? Inventory? Freight? Implementation? Service levels? Scope? Forecast guarantees? Exclusivity?

Sophisticated enterprise agreements tend to make these variables explicit. Independent professionals frequently do the opposite and reduce their entire negotiation to a single number.

A client says, “I can’t do $25,000. Can you do $15,000?” The consultant hears two possible agreements: $25,000 or $15,000. In reality, there may be twenty. Perhaps $15,000 buys a smaller scope, fewer deliverables, less access, a different timeline, no implementation, different payment terms, or a group format rather than one-to-one support.

The question does not always have to be whether you will accept less. A better question is: What does less buy?

That is an entirely different negotiation.

Subscribe now

Never Win a Deal You Cannot Afford to Deliver

BATNA taught me to understand what happened if the deal did not occur. Operating taught me to add another test: never negotiate a deal you cannot afford to lose, and never win a deal you cannot afford to deliver.

That sounds obvious until a very large revenue number is sitting in front of you.

A $5 million contract can be a terrible contract. Can you source it profitably? Can you finance the inventory? Can you meet the quality requirements? Do you have the engineering capacity? What does the working-capital requirement look like? What happens if the forecast is wrong, freight costs rise, or the customer suddenly needs twice as much product? What happens if the agreement consumes the people and capital you need for a better opportunity?

Subscribe now

Revenue is seductive. Bad revenue is still bad.

As an operator, I learned to view the agreement and the operating system required to support it as the same decision. That principle applies just as readily to a one-person consulting company. If you sell unlimited access for a fixed fee, for example, you have not merely made a pricing concession. You have designed an operating model, and it may be a terrible one.

The same experience changed my definition of what it means to win a negotiation. When you negotiate with the same customers and suppliers for years, you stop believing every transaction should be maximized independently. Some of the most valuable customer relationships we built at R.G. Brewton became many times larger than where they began. That happens only when both sides keep wanting to return to the table.

Sometimes that meant making a concession. Sometimes we invested more in engineering, inventory, service, or some new capability because we believed the relationship could become much larger. Sometimes we simply did more than the agreement technically required.

That isn’t weakness. It’s strategy.

The important distinction is whether the concession is intentional. There is an enormous difference between giving something because you believe it creates more long-term value and giving something because you are afraid the customer will say no. One is an investment. The other is fear dressed up as customer service.

My definition of a great negotiation today is therefore different from what it was when I first encountered these ideas in a classroom:

A great negotiation improves the economics, protects your ability to deliver, and increases the probability that both sides want to do the next deal.

Winning the room and losing the relationship isn’t winning. Neither is winning the contract and losing money fulfilling it.

The Operating Negotiation Framework

After enough repetitions, I found myself reducing much of what I had learned into six questions. Together, they form what I now think of as the Operating Negotiation Framework: Proof → Market → Value → Price → Terms → Walk.

It incorporates ideas I first encountered while a student at Harvard College, but the sequence reflects two decades of actually having to live with the agreements I made.

Proof asks: What can I demonstrate? Before I tell someone what something is worth, I want evidence. What have we delivered before? What outcomes have we produced? What capabilities actually exist? Why should the other person believe the promise we are making?

This has become particularly important to me now that so much of my work involves selling expertise. Your career is not merely background information. Your career is evidence. The projects matter. The promotions matter. The customers matter. The P&L responsibility matters. The failures matter. The problems people repeatedly trusted you to solve matter. You should know your proof before you ever defend your price.

Market asks: What alternatives exist? What does the market already pay? What other suppliers are available? What would it cost to hire someone, build the capability internally, buy software, engage another advisor, or simply live with the problem?

This is where Harvard’s emphasis on objective criteria and alternatives becomes particularly useful. Pricing should not begin with how confident you feel that morning. It should begin with evidence about the market surrounding the decision.

Value asks: Why does this problem matter to this buyer? A company is rarely buying the literal thing printed on the proposal. It isn’t merely buying a component, a consulting engagement, a software implementation, or an hour of somebody’s time. It is buying what happens afterward: more throughput, less downtime, lower risk, faster growth, a solved problem, an avoided mistake, or a capability the company did not previously possess.

Understanding value means understanding why the buyer is in the room at all.

Only then do I want to arrive at price. What number can the evidence actually support? Not, “What would I feel comfortable asking?” Not, “What did somebody on LinkedIn tell me consultants are supposed to charge?” And certainly not, “What number seems unlikely to scare the customer away?”

Proof. Market. Value. Then price.

For my own work today, I also want to know three prices before any meaningful negotiation begins.

There is the market price: what does this problem normally cost to solve?
There is a strategic price: is there a legitimate reason this particular relationship deserves different economics?
There is a walk-away price: below what number does the work stop making sense?

Those numbers are much easier to determine before somebody is sitting across the table waiting for your answer.

Then come terms. What else can move besides price? Scope, access, volume, timing, implementation, payment, service, commitment, deliverables, risk. Never assume that a disagreement over price has to be solved with price. Often the best agreement is hiding inside a different combination of terms.

Finally comes walk. At what point is no deal better than this deal? Know what you will not do. Know the economics that no longer work. Know the terms that create more risk than reward, and know when your capacity is worth more somewhere else.

The important thing is to determine that boundary before emotion enters the room. You should not have to invent your principles while somebody is asking you to violate them.

THE OFFER PROBLEM
For most of your career, an employer did four incredibly valuable things for you: it packaged your expertise, found the customer, established an economic model around your work, and gave you permission to defend that model.
When you become independent, all four responsibilities become yours.
Your expertise did not suddenly become less valuable. The commercial infrastructure around it disappeared.

Oper(AI)te: The Hidden Offer

Take Your Seat

Then I Put My Own Name on the Proposal

This is the part of my negotiation history I find most interesting now.

After Harvard, after running a company, after negotiating with sophisticated procurement organizations and multinational manufacturers, after more than $250 million in enterprise agreements, and after selling the company, I started working for myself.

And I became worse at negotiating.

My first coaching offer was four hours of my time for $99. One hundred people bought it in 12 days. There was obviously demand. There was also an obvious pricing problem.

The person who had spent years protecting millions of dollars of company economics had put his own name on the proposal and begun behaving completely differently.

That became more interesting as I spent hundreds of hours coaching founders, consultants, executives, and operators and watched the same pattern repeat. People who had spent 15 or 20 years confidently defending budgets, margins, compensation, contracts, and corporate pricing suddenly started apologizing for their own invoices.

They discount before anyone asks. They add another deliverable and then another. They take another call, expand the scope, and turn a defined project into unlimited access. Because they genuinely care about doing great work, they describe all of this as overdelivering.

Sometimes it is. Sometimes it is a permission problem.

The corporation gave you permission to protect its economics. Once your own name replaces the logo on the proposal, you have to give yourself the same permission.

This is also why I no longer believe pricing an offer should primarily be an act of confidence. It should be an act of research.

Instead of beginning with, “What should I charge?” start somewhere else. What have I repeatedly proven I can do? Who has an expensive reason to need it? What alternatives are they using today? What do those alternatives cost? What changes economically if the problem gets solved? Where does my experience make me unusually credible?

Then price it.

The number becomes easier to say because you are no longer asking another person to validate your self-worth. You are presenting a conclusion supported by evidence.

BUILD THE OFFER BEFORE YOU NEGOTIATE IT

This is much of what Wessal Khader and I have built into Oper(AI)te™: The Hidden Offer.
We don’t begin by inventing a product or asking what someone wants to charge. We begin with the evidence accumulated across a career. We use that evidence to identify the expertise with the greatest commercial value, map it against what buyers are paying for now, turn the intersection into one clear offer, and establish a defensible price.
Proof → Market → Value → Price.
The objective isn’t to teach an experienced operator another random skill. It is to identify the valuable skill they have already spent 15 or 20 years proving and finally build the offer around it.

Take Your Seat

The Hardest Person I Have Ever Negotiated For Is Myself

I’m grateful I got to study negotiation when I did. Harvard gave me a vocabulary for ideas I would encounter thousands of times afterward: interests, positions, options, objective criteria, alternatives, relationships, commitments.

Running companies gave those ideas consequences.

Operating taught me that price is only one part of an agreement and that a signed contract can still be a terrible deal. It taught me to create variables before giving one away, to consider the economics of fulfilling an agreement as carefully as the economics of signing it, and to recognize that the biggest opportunity is often not the first transaction but the relationship that follows it.

Entrepreneurship taught me something different.

You can spend 20 years learning to defend the value of companies, products, contracts, employees, inventory, engineering, and entire operating systems and still hesitate when the thing being valued is you.

So today I use the same framework.

Know the proof your career has accumulated. Know the market surrounding the problem. Understand the value of solving it. Let the price follow from the evidence. Create enough terms that price isn’t your only lever. Decide in advance where the agreement stops making sense.

Then be generous. Overdeliver when overdelivering is a strategic decision. Build relationships worth protecting, and create enough value that people want to sit across from you and negotiate the next deal.

And when your own name is finally on the proposal, negotiate for it with the same discipline you spent your career applying to everyone else’s.

Then send the invoice.

- j -

Subscribe now

Connect with John on LinkedIn!

The Creator Economy’s Original Sin: Building An Audience Without An Offer

John Brewton — Fri, 11 Sep 2026 14:05:14 GMT

Subscribe now

My Story

Two and a half years ago, I had a little over 500 LinkedIn connections and decided to post every day with the aim of finally building the audience had been telling me I should build for over a decade.

I studied content and built relationships with other creators. I took Justin Welsh’s LinkedIn MBA and Ship 30 for 30, then invested $5,000 in Chris Donley’s Creator Accelerator, one of the best investments I made during that period.

I learned how to grow.

My work has since generated more than 40 million LinkedIn impressions. My audience across platforms now exceeds 75,000. That includes 50,000 on my LinkedIn pages and 10,000 on TikTok. Here on Substack, Operating has 12,000 followers and 6,000 subscribers, and has become a bestselling publication.

Connect with John on LinkedIn!

For the first 18 months, I had neither a clear offer nor a clear enough purpose for the newsletter. I assumed that if I built the audience, the clients would come.

The expertise behind my best-paid work was available before any of that growth, but it was hidden within the consulting engagements that had long been my bread and butter.

My first offer was 4 hours of my coaching time for $99.

It sold +100 seats in 12 days.

I wanted to learn how to build a coaching offering. I wanted to do something crazy and unscalable that people would talk about if it succeeded. During those hours, I listened to the work and business challenges each member was facing and helped them solve the problems they brought.

Many wanted help scaling a newsletter, improving their content, or growing an audience. Most were trying to figure out what to sell and how to make money as solopreneurs. The conversation would get into the business underneath the content.

Then there were the people who already owned larger businesses. They brought 4 kinds of problems:

AI transformation.
Enterprise technology and software that needed improvement or optimization.
Sales and inventory operations strategy.
Growing EBITDA ahead of a business exit.

These were the kinds of challenges I had spent years solving in businesses I operated and companies I advised. When someone needed further help, I offered a larger coaching engagement. From there, I moved into the work I had been doing throughout my consulting career: understanding the situation, scoping the work, and building a proposal. We negotiated the engagement and decided whether to proceed.

That part was familiar. I knew how to understand a business problem and propose work to address it. I had been doing it for years.

A number of those coaching members became consulting or advisory clients. That work has brought in hundreds of thousands of dollars this year. I expect it to continue into next year.

The coaching offer itself was massively unsustainable. Four hours for $99 works out to $24.75 per coaching hour, before preparation or follow-up. More sales meant more time committed at that rate.

Those first 100 sales brought in roughly $15,000 in gross revenue. I was getting paid to learn how to build a coaching offering and whether I wanted to keep doing it. I eventually decided I did not want to build that coaching business.

The experiment also became a paid way to meet people who needed larger consulting and advisory engagements. Judging it only by the hourly coaching rate would miss most of what it produced.

Oper(AI)te™: The Hidden Offer is a four-week live build designed to turn the expertise, results, and experience you already have into an offer people will pay for. We’ll help you find it, validate the market, price it with evidence, publish the sales page, and build the AI systems that help sell it. You already did the hard part: becoming exceptionally good at something. Now we’re going to make it buyable.

Take Your Seat Today

Some people wanted advice on building an online business. Others needed the operating judgment I had developed long before I started publishing online. The conversations helped me understand the difference and decide where to put my work.

I spent my twenties studying economics at Harvard and the University of Chicago. I spent my thirties helping my father build, scale, optimize, and eventually sell an industrial distribution business. Over the last 6 years, I have consulted with manufacturing companies, B2B businesses, and technology firms.

The work I am paid best to do is operational optimization that improves financial performance. I spent 20 years learning it.

I love the creator community. Coaching solopreneurs has become some of the most rewarding work I do. I value the audience, the training, and the relationships I have built. The money I left on the table came from confusing the skill I had recently learned to distribute with the expertise I had already earned.

I could have tested an offer when I had 500 connections. I could have taken a specific problem and a proposed piece of work to people in my professional network. Their responses would have given me something to examine while I continued learning to publish.

Instead, I spent 18 months leaving the offer unresolved.

For someone with a long career, the work worth examining may be the work that feels most ordinary. Scoping a consulting engagement was familiar to me because I had done it repeatedly. Understanding a company’s operating problems was familiar for the same reason.

That familiarity made it easier to overlook the work when I was thinking about what to sell online.

Start With The Work You Have Already Done and Been Paid For

It’s time to make this mean something specifically for you. Become a paid subscriber to get the exact set of action steps and strategy I would deploy today if I was starting from zero and the workbook to help you get started.

Subscribe now

I Turned 22 Hours of Founder Office Hour Video Recordings Into a Learning Library with Claude Cowork, Opus 5, Zoom and Google Drive Connectors

John Brewton — Mon, 24 Aug 2026 11:18:48 GMT

There is a category of AI work I don’t think we talk about nearly enough. Not asking ChatGPT a question, generating an email, or summarizing a document. I mean the ugly, multi-step projects sitting on your to-do list that you know need to get done but could easily consume half a day or more of your valuable time.

I had one of those projects this week. I had 22 recordings from my Operating Founder Office Hours sessions that I wanted organized, summarized, researched, packaged, branded, linked, and ultimately turned into a learning library for my community. Doing all of that manually would have meant hours and hours of opening files, reading transcripts, finding links, researching topics, formatting documents, and organizing folders.

Instead, I gave the project to Claude Cowork and Opus 5. Across roughly four or five major prompts, my agents worked on it for about three and a half hours. I did not spend three and a half hours doing the work. Claude did the work, while I focused my enery on more important tasks and was able to spend time with my family.

While the agents worked, I kept working on other things. I responded to client emails, wrote the outline for a new article, got some reading done answer, thought through other projects, and spent time playing with my daughter. That distinction is becoming incredibly important: the value of AI isn’t simply that something that used to take three hours can now take 20 minutes, it is increasingly that a task can take an AI agent three hours while taking almost none of yours.

Watch Me Build It Here on Loom

I recorded the build as I was doing it so you could see what this kind of work actually looks like. I give Claude a set of instructions, let it work, come back to review the output, decide what the next layer should be, and then send it back to work again. The cycle was essentially direct, delegate, leave, review, redirect.

That is increasingly what agentic work looks like for me. I am still responsible for the outcome, the judgment, and the quality of the final product. But I no longer need to personally perform every action from beginning to end.

Become an Operating Founder and join our weekly live office hours this week!

Become an Operating Founder

The Project Started With 22 Zoom Calls

Every week, I hold Office Hours for the founding members of my Operating Founders community. We discuss how we are building our companies, Operating with AI, sales and go-to-market strategy, creator and content strategy, and whatever problems founders and operators are working through at the time. I record those sessions on Zoom, and over time, I have accumulated 22 recordings filled with useful conversations.

But I didn’t want a folder containing 22 recordings. I wanted someone to be able to arrive months later and immediately understand what happened in each session, watch the relevant recording, and then go deeper through additional research and resources. I wanted to turn an archive into a learning system.

Subscribe now

Prompt One: Retrieve and Organize Everything

The first layer was operational. Using my Zoom connection, I asked Claude to retrieve the recordings and associated material, move everything into Google Drive, create the appropriate folder structure, and apply a consistent naming convention.

Think about the manual alternative: finding recordings, downloading them, creating folders, renaming files, uploading everything, checking dates, and making sure every file is where it belongs. Necessary work, yes. A particularly good use of my time, no.

Prompt Two: Understand 22 Conversations

Next, I asked Claude to work through all 22 transcripts. For every session, I wanted a title, a concise description of what we discussed, the correct date, and a link back to the corresponding recording in Google Drive.

Even if I could manually review, summarize, and link each session in ten minutes, that step alone would have taken more than three and a half hours. Instead, I defined the desired output once, let Claude work through the entire collection, and returned when it was ready for review.

Prompt Three: Make the Existing Work More Valuable

I didn’t simply want summaries. For every session, I wanted my community to have high-quality resources that extended the conversation beyond what we discussed during Office Hours.

So I had Claude use an Elite Research skill I built to analyze the topics in each transcript and identify strong articles, research, media, and reference materials. The output was no longer simply “Here is what John said.” It became, “Here is what we discussed, here are the important ideas, and here are additional experts and resources that can help you explore the subject further.”

Take a look at one of the session note documents here.

Prompt Four: Turn the Work Into Finished Assets

Then I applied another reusable skill I have built around my personal brand. Claude took the transcripts, summaries, and additional research and turned them into 22 professionally formatted research and notes packages—one for every Office Hours session.

This is where the scale of agentic execution becomes easier to appreciate. Formatting documents, structuring sections, inserting research links, applying branding, naming files, exporting PDFs, and organizing the outputs are not one task. They are dozens of small tasks hiding inside a larger project, and those small actions are exactly where so much of our working day disappears.

Prompt Five: Build the Final Experience

The last step was creating a single resource that my community could actually navigate. For each session, I wanted a title, a short description, a link to the recording, a branded visual, and a link to the associated research package, all assembled into a document I could move directly into Substack.

The result was an organized library of 22 sessions rather than a pile of recordings. The overall workflow was straightforward: Capture → Organize → Understand → Enrich → Package → Publish.

Here’s a look at what the final product looks like on Substack:

Become an Operating Founder and join our weekly live office hours this week!

Become an Operating Founder

Your Agent’s Time Does Not Have to Be Your Time

This is the biggest lesson I took from the project. If someone hears that the build took three and a half hours, they might reasonably think that doesn’t sound particularly fast. But those were primarily Claude’s three and a half hours, not mine.

My time was concentrated at the places where judgment mattered: What am I trying to build? Is the output good? What should happen next? What needs to change? Once those decisions were made, I could hand the execution back and continue doing something else.

That changes the productivity question. Instead of asking only, “How much faster can AI help me perform this task?” I increasingly want to ask, “How much of this task requires me at all?”

Start Looking for Three-Hour Agent Jobs

Look through your own week for these projects. Maybe it is reviewing 30 customer calls, researching 40 companies, cleaning up a CRM, turning old webinars into training materials, analyzing proposals for common patterns, organizing a document archive, or converting sales conversations into objection-handling material.

These projects are everywhere, and many of us still calculate whether they are worth doing based on how many hours we would need to spend completing them. That calculation is becoming outdated. The better question is: Can I spend 10 or 20 minutes defining this well enough that an agent can spend the next three or five hours doing it?

You may not be able to delegate 100 percent of the work. That’s fine. Delegate 60 percent, then 80 percent, while keeping yourself at the judgment points and letting the agents own more of the execution.

Find the Work You Should Stop Doing Yourself

The bigger opportunity here is not better prompting. It is developing the ability to look at complicated work, see the complete system, break it into layers, and understand which parts require human judgment and which parts can increasingly be delegated.

Historically, I might not have built this library at all—not because it wasn’t valuable, but because the opportunity cost of spending most of a day doing it would have been too high. AI didn’t simply make the project faster. It made a project that might previously have been uneconomic worth doing.

That is what excites me most about where we are headed. The productivity revolution isn’t going to come from humans simply learning to perform the same work slightly faster with AI sitting beside them. It is going to come from learning how to orchestrate meaningful work that continues while we are doing something else.

My agents worked for three and a half hours on this project.

The important part is that I didn’t have to.

- j -

Join the Operating Founder Office Hours each week to work through the real problems, decisions, and opportunities you are facing alongside other founders and operators. Every session is recorded and added to a growing learning library of summaries, research, and additional resources, so the value of the conversation continues long after the call ends. Come for the live discussion, and leave with a deeper body of knowledge you can keep using.

Become an Operating Founder

John Brewton documents the history and future of operating companies at Operating by John Brewton. He is a graduate of Harvard University and began his career as a PhD student in economics at the University of Chicago. After selling his family’s B2B industrial distribution company in 2021, he has been helping business owners, founders, and investors optimize their operations ever since. He is a career consultant and business operator, asking the question: What is the future of companies?

The Infinite Value of a Meeting Transcript for the AI-Fluent Operator: How to 100X the Value of Meeting Transcripts in Your Workflows and Business

John Brewton — Tue, 18 Aug 2026 10:22:18 GMT

Subscribe now

Most companies are using AI to solve the least interesting problem with meetings.

They record the call. They generate a transcript. They ask AI for a summary. Someone skims it, closes the window, and the team moves on with their business.

That is useful. But it dramatically underutilizes the opportunity at hand. A meeting transcript is not merely a record of a conversation. It is a structured source of decisions, commitments, customer intelligence, objections, deadlines, requirements, ideas, risks, owners, and institutional knowledge.

In an AI-NAtive company, all of those things can become inputs into other systems.

The transcript can update the CRM. It can create tasks. It can draft the follow-up email. It can write the project brief. It can generate a proposal. It can prepare an executive summary. It can tell someone who missed the meeting exactly what they need to know. Months later, an AI agent can search thousands of previous conversations and answer a question nobody remembers discussing.

That changes the purpose of the transcript entirely.

The transcript should not simply tell us what happened in the meeting. It should reduce the probability that another meeting is required.

That may eventually become one of the most important operating advantages of AI.

A Practical Example: Turning One Transcript Into a Finished Asset

Watch the Loom video here.

I wanted to show one small example of what this looks like in practice.

Every Friday, I host Operating Founder Office Hours for members of my paid Substack community. We spend the session working through questions from operators, builders, solopreneurs, and small-business owners. Those conversations can move across dozens of ideas, resources, frameworks, and problems in a single hour.

Historically, the recording itself would have been the primary asset.

Now the transcript becomes the starting point.

In the Loom above, I show Claude pulling the recording and transcript directly from Zoom and creating a comprehensive, anonymized set of session notes. I then have it run my Elite Research skill across the conversation, identifying the concepts, books, academic research, articles, and thought leadership connected to what we discussed.

That produces an expanded research library members can use to keep learning long after the call ends.

Finally, I use my JB Personal Brand skill to turn everything into a polished PDF that can be distributed alongside the recording.

One conversation becomes a summary, a knowledge resource, a research library, a learning tool, and a finished community asset without requiring me to manually reconstruct the session.

Members can then feed the document back into Claude or ChatGPT, ask which resources are most relevant to them, explore an individual topic, or have an extended conversation about the research.

Record it once. Structure it once.

Reuse the intelligence many times.

The same principle becomes much more powerful when applied across an entire company. Here is a sample of the weekly session notes that all Operating Founders receive from our weekly calls:

Subscribe now

What You’ll Learn in This Article

Why is the meeting transcript becoming an operating layer rather than just a record of what happened?
How transcripts can automatically improve CRM quality by capturing decisions, objections, commitments, stakeholders, and next steps.
How AI can turn meetings into execution by creating tasks, assigning owners, identifying deadlines, and surfacing blockers.
How better transcript workflows can eliminate future meetings, especially recap, handoff, status, and information-sharing calls.
How one conversation can become an asset factory, producing follow-ups, briefs, proposals, project plans, and internal documentation.
Why thousands of transcripts can become proprietary institutional intelligence that teams and AI agents can search and learn from.
The new KPI for meetings: not how well they were summarized, but how much work they created—and how many future meetings they made unnecessary.

What You’ll Get With This Article

The Meeting Transcript Operating System: Paid subscribers get a complete interactive workbook designed to be uploaded directly into Claude or ChatGPT. Bring one real meeting transcript, and the workbook walks you through building a repeatable system for extracting decisions and commitments, routing intelligence into your CRM and other tools, creating high-value assets automatically, identifying meetings that can be eliminated, and turning the finished workflow into reusable AI instructions, skills, scheduled tasks, and agents. You leave with a working Transcript-to-System Map, CRM Extraction Schema, Asset Factory, Meeting Elimination Report, Transcript Agent Instruction Set, Transcript Value Scorecard, and a 30-day implementation plan. This is not simply another set of prompts.

Become a Paid Subscriber

How I built a five-member Excel team in Claude and how you can too.

John Brewton — Wed, 05 Aug 2026 12:11:45 GMT

The best AI model in one financial-modeling benchmark edits 89.69 percent of the correct cells.

It gets only 34 percent of the final answers right.

That gap explains the problem with most AI spreadsheet workflows.

The model can make a lot of plausible-looking changes. It can move quickly. It can even appear productive. But none of that guarantees that the final answer is correct, traceable, or safe to use.

That is why I spent this morning building a 17-page spreadsheet operating system instead of writing another “better” prompt.

I did not build one all-purpose spreadsheet agent.

I built a five-role team.

Here is how I got there.

Watch my live loom recording of the process here.

First, Understand What the Workbook Is

Before I explain the research or introduce the five roles, it is worth explaining what I mean by a workbook.

This is not a workbook in the traditional sense.

It is not a document you print, complete by hand, and place in a binder. It is also not simply a collection of prompts.

The workbook is a system of instructions that the AI reads and then uses to interview you. You upload it into a new working session and ask the model to walk you through it. The model follows the rules inside the workbook, asks you a structured series of questions, and collects the information it needs to build a working AI team around the way you actually operate.

That distinction matters.

Most generic AI workflows begin with a finished prompt. Someone else decides how the work should be done, packages those assumptions into a template, and hands it to you.

The workbook begins one step earlier.

It helps the model learn how you need the work to be done before it starts building anything. For spreadsheet work, that means collecting details such as:

The types of analysis you perform
The workbooks and files you regularly use
The outputs you need to produce
The formulas and modeling standards you expect
The software and data sources involved
The assumptions that must always be stated
The controls required before work can be published
The recurring tasks that may eventually be scheduled

The result is not one enormous prompt that tries to anticipate every possible spreadsheet request.

The result is the infrastructure needed to handle many different requests consistently.

Why the Workbook Is Built Around Questions

The quality of an AI system depends heavily on the quality of the context behind it.

But most people do not naturally provide that context in one clean, complete prompt.

They leave out details because those details feel obvious to them. They forget to explain how a file is organized, which source should be treated as authoritative, what level of precision is acceptable, or what must happen before an answer can be shared.

Those omissions become failures later.

The workbook is designed to prevent that by turning system design into a guided conversation.

Instead of asking you to sit down and write a complete operating manual from scratch, the model asks one question at a time. Your answers become the raw material for the skills, role instructions, rules, checklists, and scheduled tasks that the team will use.

This makes the process more accessible, but it also makes it more rigorous.

The questions force decisions that a loose prompt allows you to avoid.

What is the team responsible for?
What is outside its authority?
Which files must it inspect?
Which software must it use?
What must it produce before another role can begin?
What conditions should stop the workflow?
How should the final work be checked?

By the end, the model is no longer guessing what “good spreadsheet work” means to you. You have defined it.

Why the Instructions Live in a Workbook

I use a workbook because the system needs to be portable, repeatable, and visible.

A good conversation with an AI can produce excellent work once. But conversations are easy to lose, difficult to audit, and hard to reproduce with another person or in another session.

The workbook gives the process a durable structure.

It contains the questions to ask, the order in which to ask them, the rules that must be followed, and the outputs that should be created from the answers.

That means I can upload the same workbook into a clean session and have the model run the process again without relying on the memory of a previous conversation.

It also means I can give the workbook to someone else.

They do not need to understand every decision I made while designing it. They can upload it, ask the model to follow it, answer the questions, and build a version of the system that reflects their own work.

The workbook is therefore doing two jobs.

First, it captures the operating logic behind the system.

Second, it teaches the user how to build that system for themselves.

Why the Workbook Comes Before the Team

The five roles I built are important, but the roles are not the starting point.

The workbook is.

Before the AI can build an Intake Clerk, Analyst, Modeler, Auditor, or Publisher, it needs to understand what those roles mean inside your environment.

Your Analyst may need to prioritize cash flow, while someone else’s Analyst focuses on inventory or customer acquisition.

Your Modeler may work inside Excel with files stored in SharePoint. Another company may use Google Sheets, a finance platform, and data exported from its ERP.

Your Auditor may need to reconcile every result to the general ledger. Someone else may need to compare it with a prior board report or an external data source.

The role names can remain the same while the actual instructions change substantially.

The workbook collects the information needed to make those roles specific.

It turns a general team design into your team design.

That is why I do not begin by telling the model, “Build me five spreadsheet agents.”

I begin by giving it a structured process for learning what those agents must know, what they are allowed to do, what they must never do, and how their work should move from one role to the next.

The Workbook Is the Control Layer

A useful way to think about the workbook is as the control layer between you and the AI team.

The roles perform the work.

The skills give the roles specialized methods.

The connected software, files, and data give them access to the materials they need.

The scheduled tasks determine when recurring work happens.

But the workbook defines how all of those pieces should be assembled.

It creates the instructions that govern the system before the system begins operating.

That is why I have started building workbooks for more and more types of work. They give people a repeatable way to create new AI capacity without needing me to personally design every team, skill, or workflow for them.

Once you understand how to build one, you can use the same approach for inbox management, research, financial analysis, reporting, project management, or almost any other structured knowledge-work process.

You are no longer collecting isolated prompts.

You are learning how to design the team that will use them.

Now for the process I documented for you today…

Phase 1: Research the failures

Time: 40 minutes

I began with one question:

What actually goes wrong when AI works inside a spreadsheet?

I limited the research to benchmarks and spreadsheet audits. No opinions. No productivity claims. Just documented failure modes.

Three findings shaped the entire system.

1. AI often fails because it does not inspect the workbook properly

SpreadsheetBench 2 evaluates AI agents on multi-sheet workbooks with real formulas, dependencies, and linked calculations.

The best model achieved 34.89 percent overall accuracy.

On debugging tasks, it scored just 12 percent.

The researchers identified two recurring problems:

Insufficient inspection
Incorrect target selection

In plain English, the models often failed before they started calculating.

They did not fully understand the workbook, or they changed the wrong cells.

That is not primarily a math problem. It is a looking problem.

2. AI can produce the right number in the wrong way

WorkstreamBench evaluates agents on end-to-end financial tasks.

The best agent scored 69.1 out of 100. Once the task required several chained calculations, the score fell to 53.4.

One of the most important documented failures was deceptively simple: agents sometimes pasted a correct number into a cell where a formula should have been.

The answer may be right today.

But the workbook is now broken.

Change an assumption next week, and the pasted number will not update. A reviewer cannot follow the calculation. The spreadsheet becomes a static answer rather than a working model.

Would you like to work with John?

Click Here

3. Review cannot be treated as an afterthought

Raymond Panko’s field audits found errors in 91 percent of operational spreadsheets.

A single reviewer typically catches about half of the errors.

A second reviewer, working independently and using a checklist, can raise detection to roughly 80 percent.

That matters because most AI spreadsheet workflows ask one agent to inspect the data, choose the method, build the model, calculate the result, and verify its own work.

That is the spreadsheet equivalent of asking an analyst to approve their own analysis.

Subscribe now

Phase 2: Design the team

Time: 20 minutes

The research killed my original idea.

I had planned to build one spreadsheet employee, similar to the inbox employee I taught last month.

But the evidence pointed in another direction.

Inspection, analysis, modeling, and verification are not one job. They are separate jobs with different failure modes.

Combining them inside one large prompt does not create a capable employee. It creates an analyst checking their own work.

So I divided the workflow into five roles.

Each role has:

One primary responsibility
One artifact it must hand to the next role
One action it is not allowed to take

The Intake Clerk

The Intake Clerk profiles the workbook before any analysis begins.

It identifies the sheets, tables, formulas, inputs, outputs, date ranges, missing fields, and potential data-quality problems.

It is not allowed to calculate the answer.

Its job is to understand the environment before anyone starts changing it.

The Analyst

The Analyst decides how the question should be answered.

It defines the method, assumptions, required calculations, and level of precision before producing a result.

The rule is simple:

State the method before stating the number.

The Modeler

The Modeler turns the approved method into a live workbook.

It builds formulas, links assumptions to outputs, and creates a model that recalculates when the inputs change.

It is not allowed to hardcode a value into a computed column.

If a number is calculated, there must be a formula behind it.

The Auditor

The Auditor independently re-derives the headline number using a second method.

It checks the formulas, assumptions, units, ranges, and reconciliation gaps.

It also has the authority to stop the workflow.

If the analysis does not hold, the Auditor issues a hold.

The Publisher

The Publisher prepares the finished workbook and communicates the result.

But it cannot publish over an audit hold.

That rule is what turns the system into a team rather than five prompts pasted one after another.

Every role can reject the work it receives.

Phase 3: Build the operating system

Time: 45 minutes

The finished workbook is 17 pages.

It includes:

Nine questions that define your analysis standards
Eight prohibited spreadsheet habits
Four prompts to run in sequence
Six reusable prompts for future analyses

The prohibited habits may be the most important part.

They include:

Pasting values into cells that should contain formulas
Reporting numbers with no supporting cell reference
Applying assumptions without stating them
Presenting more precision than the underlying data supports
Editing a workbook before profiling its structure
Publishing an answer that has not been independently reconciled

Each prohibition is paired with a replacement rule.

That distinction matters.

Telling an AI what not to do is rarely enough. You also have to define what it should do instead.

And every rule in the workbook traces back to a documented failure mode. These are not formatting preferences. They are controls.

What this system actually gives you

Most people use AI in spreadsheets because they want speed.

But speed is not the main constraint.

A fast answer with no formula trail still has to be checked by hand. That manual review is the work you were trying to eliminate in the first place.

The real opportunity is not faster analysis.

It is additional analytical capacity.

This system gives an operator access to three functions that many small teams have never been able to staff properly.

A data profiler

Someone examines the workbook before the analysis begins.

They identify what is present, what is missing, what is linked, and what may be unreliable.

A modeler

Someone builds a workbook that continues working when the assumptions change.

The goal is not a spreadsheet that was correct once. It is a model that remains useful.

An auditor

Someone independently checks whether the headline number holds.

This is the role most small finance teams do not have. It is also the role the research suggests matters most.

A private company with a four-person finance team may never have had a second analyst independently re-derive a number before it reached the board.

Historically, that verification could cost as much as the original analysis.

It does not have to anymore.

The final output is not simply a number.

It is:

The number.
The method above it.
The formulas beneath it.
And an independent check beside it.

What to try this week

Choose one analysis you completed manually last month.

Pick something where you already know the correct answer.

Then run it through the five roles in order.

Have the Intake Clerk map the workbook before making any changes.

Have the Analyst explain the method before calculating the result.

Have the Modeler prove that the headline cell contains a formula rather than a pasted value.

Have the Auditor calculate the answer a second way and report the difference between the two results.

Every place where the AI’s work differs from your manual process should become a new rule in your operating profile.

Repeat that process for a week, and the system becomes more consistent.

Then use the Auditor on the spreadsheets your team produced last quarter.

Panko found errors in 91 percent of operational spreadsheets.

There is no reason to assume ours are the exception.

Become a paid subscriber to receive the working and build your Excel team this week!

Subscribe now

And if building this kind of AI fluency is especially appealing to you, and you want to learn everything you can in the year ahead, become an Operating Founder and join my 52-week-per-year Friday live Q&A sessions with The Operating Founder Community. We are a +200-strong group of Operators and Builders choosing a new vision for our futures.

Become An Operating Founder

Download the workbook to build your Excel team below.

Stop Arguing (or Caring) About AI Writing About Slop or Not Slop

John Brewton — Sun, 26 Jul 2026 16:20:58 GMT

Historically, every time a tool made ideas cheaper to share, the people who had mastered the older, slower way treated it as a threat to defend against rather than a shift to move with. The scribes who spent their lives copying manuscripts did not celebrate the printing press. They saw a craft and a livelihood going to zero. Plato warned that writing itself would ruin memory and hollow out real understanding.

That was around 2,400 years ago.

Civilization grew anyway.

We are running the same fear-based reflex again. Our social feeds and inboxes are full of people litigating whether AI writing counts, whether it is cheating, whether the craft is dying. The scribes and the press are at it again. And while this argument burns, the enabling machines that create distribution for our ideas, work, products, and services are moving faster than they have in 500 years.

Here’s our current reality: The cost of producing words just fell to zero, and the only things left worth owning are the idea you choose and the distance it travels.

Subscribe now

Five thousand years of the same job

For 5,000 years, every writing tool did one thing successfully.

It made ideas cheaper to share.

Clay tablets held temple receipts, not poetry.

A medieval book took weeks of skilled labor and got chained to the desk because it was worth more than the room it sat in.

Then the printing press cut the price of a book by two-thirds within 50 years.

Each drop did the same work.

More people could read, and more people could write.

This cycle of payoffs compounded for centuries.

In 1820, 12 percent of the world could read. Today it is roughly 90 percent. Cheaper copies did not make people read less. This innovation turned reading from an elite privilege into a basic human skill.

Every one of those tools shared a blind spot. The personal computer and the laptop made producing and editing words faster, and none of it touched the oldest cost of all, the effort of composing the thought itself. You still had to sit down and produce the words. That is the cost ChatGPT and Claude just took to zero. These are the first tools that compose and reason back. They draft, they read, they research across sources in minutes instead of weeks.

Subscribe now

What actually got scarce

When a cost falls to zero, value moves to whatever sits next to it and is still expensive.

Composition is now cheap. So look at what it sits between. Upstream is the choice of what is worth saying. Downstream is getting it to the right people and being believed once it lands.

Judgment and distribution.

Those are the survivors.

Everything the AI-writing debate is fighting over is the one output that now fails to be scarce. The question was never how fast you can write. Speed is solved. The new question we should be centering our collective, ambitious, entrepreneurial gaze on is what you choose to think about, and how far you can move it.

The gate is gone

Here is my take on the opportunity we should all be aligned behind, but very few actually are.

For most of the internet’s life, reach was a function of accumulation. You built a list, a following, a set of contacts, and your ideas reached roughly the people you had already collected. That is the social graph. If you were small or new, you stayed small. Distribution was gated by who you already knew.

That gate is gone. A recommendation model now reads the content itself, scores it against what each person actually engages with, and serves it to strangers who have never heard your name. Follower count stopped being the gate and the metric to chase. The quality of your idea and what you can use the new stack to produce, distribute and build with it is the new hurdle, the new gate.

This is not a forecast I’m making as some Substack author you’ve never heard of befroe. Meta told a federal court, in its own antitrust filing, that just 7 percent of the time people spend on Instagram, and 17 percent on Facebook, is spent on content from friends. Zuckerberg has said more than 50 percent of what people see on Instagram is now chosen by AI, not by who they follow. The man who built the social graph testified that Facebook is no longer the culture, and floated wiping everyone’s friend connections to reset the platform around interests. The social network is barely social. The algorithm runs the room.

Subscribe now

Social media is dead. AI-enabled, interest-based media is the replacement.

So yes, the interest graph removed the gate, but it did not remove the competition. Cheap distribution does not spread attention evenly. Anita Elberse at Harvard studied this directly. In one set of 8 million songs that each sold at least 1 copy, roughly a third sold exactly 1 copy. Attention still stacks on whatever earns it. A model serving your idea to strangers is a momentary shot, not a guarantee, by far.

So the constraint did not disappear. It has just moved to a new place in the supply chain. It used to sit on permission, on access, on the size of your list. Now it sits entirely on the quality and clarity of the idea and how well you shape it. That is a harder game to fake and a fairer one to win. For the first time, a small company with a sharper idea can out-reach an incumbent with a giant list, because the list no longer decides.

Voice is the accelerant

If the interest graph decides who gets heard, voice decides how fast you can feed it.

Voice is the most information-dense form of human communication, and for the first time it is programmable. Andreessen Horowitz calls it the first and primary way people will now interact with AI, and 22 percent of the most recent Y Combinator class were voice-agent companies. You speak, the machine drafts and researches, and the loop from raw thought to finished asset closes to near real time. The cost of running these models fell about 280 times in 18 months, so the constraint is no longer money or minutes. It is what you point them at.

The reading side moved the same way. Most Google searches now end with no click at all. People take the answer, not the source. The idea that the machine surfaces matters more than owning the page it came from.

What to do with this

The uncomfortable part is that none of this rewards the skill most knowledge workers spent 20 years building. Fast, clean production is now table stakes a machine clears for free. What it rewards instead is judgment, and judgment is harder to fake.

So we need to focus our operating energies on the part that pays.

First, spend your scarce hours choosing ideas, not producing words. Keep a running list of the few ideas you want to be known for. That choice is now the work that pays most.
Second, move at conversation speed. Talk your thinking out loud, let the model draft and research, then you cut and judge. Do not type what you can speak.
Third, publish into the interest graph on purpose. Turn each idea into something consumable and teachable, ship it on a schedule, watch what the algorithm carries, and make more of that.
Fourth, own a few ideas in public until your name is attached to them. Repetition around a tight set of ideas is how you get known for them. Do not scatter.
Fifth, convert reach into revenue. Distribution is not a vanity metric. The point is to become known, sell more product, and sell more service off the back of the ideas.

Let other people spend the decade arguing about whether how the idea was written or produced matters. Don’t want to give away the ending, but none of that is going to matter one bit. Pick 1 idea worth being known for and publish it every week for the next 52. The writing was never the asset. What you choose to think about, create from those ideas, and how far you can push those concepts into the world of products, serrvices, subscriptions, VALUE FOR YOUR AUDIENCE, is the only thing that matters.

Stop worrying about debates that don’t matter.

Stop letting irrelevant online voices shame you.

Get to work, Substack.

- j -

The Interest-Graph Operating System

I built the system that is the logical, operating approach for this set of ideas. And my paid subscribers get it. The Interest-Graph Operating System is a workbook you paste into your own model (anyone you like or prefer), which then turns into your facilitator and walks you through the four moves: choosing the ideas worth owning, composing them at conversation speed, publishing them into the interest graph, and turning reach into revenue.

You answer its questions in your own words. It compiles the answers into a system you can run this week: your idea ledger, your prompt pack, your publishing plan, and the four key numbers that tell you if it is working.

If you’re just starting to understand these ideas, this workbook is the place to begin building the AI-fluent machine to architect your new professional future and business ambitions.

Become a Paid Operating Subscriber

DOWNLOAD BELOW AFTER THE PAYWALL, FOLLOWING THE APPENDIX AND FAQ SECTION.

Appendix: Sources

Opening and the history of writing tools

Plato, Phaedrus (Socrates on writing and the loss of memory), Perseus Digital Library, Tufts University. https://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:1999.01.0174:text%3DPhaedrus:page%3D275
Jeremiah Dittmar, “Information Technology and Economic Change: The Impact of the Printing Press,” Quarterly Journal of Economics, 2011 (book prices fell by roughly two-thirds; cities that adopted the press grew about 60 percent faster). https://academic.oup.com/qje/article-abstract/126/3/1133/1855353
Jeremiah Dittmar, plain-language summary, CEPR / VoxEU. https://cepr.org/voxeu/columns/information-technology-and-economic-change-impact-printing-press
“Literacy,” Our World in Data (global literacy 12 percent in 1820 to roughly 90 percent today). https://ourworldindata.org/literacy

The distribution shift: social graph to interest graph

“Facebook is Dead; Long Live Meta,” Stratechery by Ben Thompson, 2025, citing Meta’s FTC antitrust filing (7 percent of Instagram time and 17 percent of Facebook time spent on content from friends). https://stratechery.com/2025/meta-earnings-meta-turns-the-dial-social-network-r-i-p/
Mark Zuckerberg, Meta Q1 2024 earnings call, reported by Benzinga (more than 50 percent of Instagram content is now AI-recommended). https://www.benzinga.com/news/24/04/38420928/meta-ceo-mark-zuckerberg-more-than-50-of-the-content-people-see-on-instagram-is-now-ai-recommended
“The AI behind unconnected content recommendations on Facebook and Instagram,” Meta AI, 2023 (recommendation of content from accounts users do not follow). https://ai.meta.com/blog/ai-unconnected-content-recommendations-facebook-instagram/
“Spinning off Instagram, the decline of friending, and other takeaways from Mark Zuckerberg at the FTC monopoly trial,” CNN Business, April 16, 2025. https://www.cnn.com/2025/04/16/tech/mark-zuckerberg-testimony-meta-ftc-trial/index.html
Anita Elberse, Blockbusters, Harvard Business School, via “The Way of the Blockbuster,” Harvard Magazine, 2013 (market concentration; a third of 8 million songs sold exactly 1 copy). https://www.harvardmagazine.com/2013/12/the-way-of-the-blockbuster

Voice and the reading side

“AI Voice Agents: 2025 Update,” Andreessen Horowitz (voice as the first and primary interface; 22 percent of the latest Y Combinator cohort). https://a16z.com/ai-voice-agents-2025-update/
“The 2025 AI Index Report,” Stanford Institute for Human-Centered AI (inference cost fell roughly 280x in about 18 months). https://hai.stanford.edu/ai-index/2025-ai-index-report
“Google zero-click searches study,” Search Engine Land, 2026 (most Google searches end without a click). https://searchengineland.com/google-zero-click-searches-2026-study-479717

Frequently Asked Questions

On “Stop Arguing About AI Writing” and the interest-graph shift

Isn’t most AI writing just slop? Some of it is. Slop is a production problem, and production is the part that got cheap. The point of the piece is that arguing about whether a machine drafts well is the wrong fight, because the value moved to the two things a machine does not do for you: choosing the idea and judging what comes back. Quality still matters. It just stopped being about who typed the words.

Does this mean writing skill no longer matters? Writing skill matters less as production and more as judgment. Fast, clean production is now table stakes a model clears for free. What pays is taste, the ability to choose what is worth saying and to cut what is not. That is a harder skill than typing, and it is the one worth building.

What is the interest graph, in plain terms? The social graph reaches the people you already follow or already know. The interest graph reaches strangers, because a recommendation model reads your content and serves it to whoever is likely to care. Follower count stops being the gate and the idea becomes the gate. Meta told a federal court that only 7 percent of time on Instagram, and 17 percent on Facebook, is now spent on content from friends.

I have a small audience. Can this actually work for me? Yes, and that is the opportunity. The interest graph does not ask how big you are. It asks how good the idea is. A sharper idea from an unknown account can out-reach an incumbent with a large list, because the list no longer decides who gets seen.

If everyone can produce, isn’t the competition infinite? The competition is real, and attention still concentrates on a few winners. Anita Elberse at Harvard showed that digital markets are more concentrated, not less. The gate moved. It did not vanish. Removing the permission barrier does not hand you the crowd. It gives you a shot you did not have before, and the idea is what earns it.

Isn’t building on someone else’s algorithm risky? It is a real risk, and it is the cost of the reach. Manage it two ways. Convert algorithmic reach into assets you own, an email list and paying customers, so a platform change cannot erase you. And do not depend on one platform. Publish where the idea travels and move with it.

Do I need to use voice and AI to do this? No, but they are the accelerant. Voice is the most information-dense way to get an idea out of your head, and the model turns it into a draft you judge and cut. The loop from raw thought to finished asset closes to minutes. You can do the work by typing. You will just be slower.

How do I make money from this, not just collect reach? Attach an offer to every idea you choose to own. Publish the argument free so it travels, and gate the system, the template, or the service that acts on it. Track four numbers every week: reach, email subscribers, paid subscribers, and revenue. If a week moves none of them, change the idea, not the effort.

How is this different from telling me to post more? Volume without a chosen idea trains the algorithm on nothing and trains the reader on nothing. This is the opposite. Own 3 to 5 ideas, repeat them in new forms until your name attaches to them, and let the schedule do the compounding. One flagship a week for 52 weeks beats 12 brilliant pieces shipped on mood.

How much time does the weekly system take? The cadence is one flagship and five short cuts a week. Monday you choose the idea, Tuesday you compose with the prompt pack, Wednesday you ship, and the rest of the week you distribute and log the numbers. The initial build takes about 40 minutes. The upkeep is a few focused hours a week.

What is the Interest-Graph Operating System, and how do I use it? It is a workbook you paste into your own model. The model becomes your facilitator, asks you one question at a time, and compiles your answers into a finished system: your idea ledger, your prompt pack, your publishing plan, and your scoreboard. You build it once, then you run it every week. Subscribers get it with this piece.

How to Produce World-Class Research with Perplexity Computer & Claude -> Fable 5

John Brewton — Sun, 19 Jul 2026 18:12:58 GMT

The full methodology and workflow behind a 22-page, 12-chart research report: Perplexity Computer for data collection and essential research, then Claude on Opus 4.8 and Fable 5 for verification and production, along with my elite research skill file that holds the whole thing to an institutional standard. The skill’s architecture is broken down below, and the workbooks for building your own are attached.

Too Long →Didn’t Read

Perplexity Computer produced a 22-page research report and a full data workbook in 33 minutes across roughly 140 agent tasks. It still contained 5 substantive errors.
The core doctrine: never publish research from inside the machine that produced it. Move the output to a second, different model and audit it against a written standard.
The audit runs on a skill file with 6 sections: a closed 3-tier source architecture, 4 research modes, search and paywall discipline, 3 output templates, voice rules, and 8 quality checks.
Charts get rebuilt from the raw data in your own visual identity. The final report is written in your own voice and read by you in full before anything ships.
Attached to this post: the full report this article walks through, plus the Your Research Employee workbooks in 4 editions (Claude, ChatGPT, Gemini, Perplexity) so you can build your own analyst in about an hour.

The report runs 22 pages.

It carries 12 original charts, 4 parts, 2 appendices, and a complete source register with a live URL behind every figure.

The machine research behind it took 33 minutes across roughly 140 tasks.

The verification pass that followed caught 5 errors before a single chart was drawn.

The complete process took my team of agents 65 minutes. The work would have taken me two weeks.

Anyone can get an AI research report in 2026. The tools are extraordinary and improving quarter over quarter. Which means the differentiation has moved. World-class research is no longer about access to information. It is about the discipline you wrap around the machines and agentic teams you deploy to make it happen.

This piece walks through the exact pipeline I used to produce “The Last Bubble Went Public. This One Is Staying Private,” my quantitative history comparing the dot-com bubble of 1995 to 2003 against the AI boom of 2022 to 2026. I recorded the whole build on Loom. This is the written version, plus the part the video could not show: the full anatomy of the elite research skill that runs quality control, and the workbooks for building a research employee of your own on Claude, ChatGPT, Gemini, or Perplexity.

Hope this helps you this week, friends!

- j -

If you want to become AI-fluent this summer and learn how to build a team of agents to do the work discussed in this article and so much more, you should seriously consider joining my Oper(AI)tors Summer of AI Fluency cohort.

Grab Your Seat Now!

Use code TAKE500 to receive a $500 discount on your seat.

What is the best AI stack for producing research?

Perplexity Computer collects. Claude verifies and produces the output in my primary stack. Skills encode the standards so neither step depends on my memory or my mood.

Three tools, three jobs, no overlap.

The research ran in Perplexity’s Comet browser (download today!), which is my daily browser on every machine I own. The verification, chart production, and final write-up ran in Claude Cowork sessions on Opus 4.8 and the Fable 5 model. Dictation ran through Wispr Flow, which is why I can commission a research project at the speed of speech.

Step 1: How do you brief an AI research agent like Perplexity Computer?

I opened Perplexity Computer and dictated a brief. Not a query. A brief. The distinction decides the quality of everything downstream.

Watch the Loom here.

The brief had 4 components:

The history: Deep research on the dot-com bubble and crash, from the late 1990s through the early 2000s. I want to understand the period, not skim it.
The comparison: Set that period against the current AI cycle, quantitatively and qualitatively. Startups then against startups now. Trajectories then against trajectories now.
The data demand: Come back with financial data and historical data sufficient to build charts, graphs, and visualizations. This clause is the one most people skip. If you do not ask for chartable series, you get prose, and prose cannot be re-plotted in your own brand.
The standard: Exhaustive, attributable, accurate. Every claim is sourced.

Then I left my computer for a bit, and Perplexity deployed a team of agents that worked for 33 minutes across roughly 140 discrete tasks. Watching the full run is instructive, partly because Perplexity shows which models it selects for which activities. But you do not need to watch. You need the 2 deliverables it returns: a complete quantitative and qualitative report, and an Excel workbook holding every underlying data series as chartable tables.

View the Perplexity Excel data file here.

That workbook is the asset. The report is a draft. Hold that hierarchy in your head, because the next step depends on it.

Subscribe now

Step 2: How do you verify AI research for accuracy?

Here is the doctrine at the center of this whole system. Never publish research from inside the machine that produced it.

Perplexity Computer is a serious research engine. The output is attributable, the writing is clean, the coverage is broad. And it still shipped me 5 errors. Not typos. Substantive, number-level errors of the kind that end up in someone’s investor memo and then in someone’s screenshot.

So the first thing I do with any Perplexity research package is take it out of Perplexity. I downloaded both files, opened a Claude Cowork session, attached them, and ran my elite research skill against the full package with one instruction: check that everything here is accurate, and show me what is not.

The audit came back (read the full audit MD file here) with a list.

The 5 that mattered most

Cisco’s crash was understated by 22 percentage points: The figure that circulates for Cisco’s dot-com decline is 66 percent. That number measures only through December 2000, which is 22 months before the actual bottom. The true peak-to-trough is 88 percent, from March 2000 to October 2002, from $79 to $9.50. The distinction changes the lesson. Being right about the technology protected Cisco’s business. It did not protect the price.
Anthropic’s valuation ladder was mislabeled: The verified sequence reads $61.5 billion, $183 billion, roughly $350 billion, then $965 billion, with the $965 billion Series H dated May 28, 2026. Versions dating that figure to November 2025 conflate it with the Microsoft-Nvidia deal, valued at roughly $350 billion. A 15x repricing in 14 months deserves correct dates.
The AI venture funding series stopped a year early: The draft ended the series at 2024, at $114 billion. The full record runs through 2025 at $202.3 billion, close to half of all venture capital deployed worldwide that year. Comparisons ending at 2024 understate the cycle. Flattering the past is the most common failure mode in bubble analysis, and it usually arrives disguised as a data cutoff.
The marquee AI IPO count was stale: The correct count is 2. Cerebras was listed on the Nasdaq on May 14, 2026, joining CoreWeave’s March 2025 listing.
The NASDAQ year-end close needed pinning: The verified 2025 monthly series puts it at 23,241.99.

I read the list, approved all 5 corrections, and instructed that all work going forward be updated against them. Every downstream chart and paragraph automatically inherited the corrected numbers. The final report states each correction openly in its methodology appendix, which is what a reader deserves and almost no AI-assisted research bothers to do.

The general principle: collect everything in one place, then verify with a different model, ideally a stronger one than the one that did the collection. Two models will not share the same blind spots. One model grading its own homework will.

This is the first layer of checking, not the last. When the full report is assembled, I read the entire thing myself. The machine audit removes the errors a human skims past. The human reader catches what no rule can encode. You need both, in that order.

Step 3: How do you rebrand AI-generated charts in your own visual identity?

The research you publish should look like yours. Perplexity’s charts are competent and generic. So the next instruction to Claude: take my personal brand skill, take every data series from the workbook, and re-create every chart plus any additional visualizations the data supports.

Claude returned 12 charts. Each one is rendered in my visual identity, the Forest Ink field register, as an individual PNG, plus a single PDF holding all 12, plus an HTML index page for reviewing the full set in one scroll. NASDAQ’s 6.2x five-year multiple. The casualty table shows that the survivors still fell to 88 percent. The private-marks ladder no public index can show you. The 280-versus-2 IPO chart encapsulates the entire thesis in a single image.

View all twelve charts here.

The proportion and consistency across the set is the point. A reader cannot articulate why a report feels institutional. Uniform chart grammar is most of the answer.

Two practical notes. First, this only works because the data came back as chartable tables in Step 1. Second, keep the individual PNGs. You will reuse single charts in newsletters, decks, and posts for months. The report is one container for the charts, not the only one.

Step 4: How do you turn AI research into a publishable report?

Final instruction: produce the complete research report, including the charts, written in the Operating voice. Clearly written, professionally written, minimal emotion. Tell the story, walk the reader through the numbers, let the two periods of history do the arguing. Return it as a branded PDF.

It came back as a 22-page document: executive summary, 4 parts, a methodology appendix listing the corrections, and a full source register. Some charts embedded in the first PDF pass imperfectly. This does not bother me. I had Claude produce a Markdown version I could format and embed cleanly myself. Expect this. The last 5 percent of production is still manual, and pretending otherwise is how polished-looking errors ship.

Then the human read. The whole report, my eyes, before anything publishes. The machines compress 2 weeks of analyst work into an afternoon. They do not remove the afternoon.

The finished report. 22 pages, 12 charts, 4 parts, every correction stated in Appendix A, the complete source register in Appendix B.

What goes inside an elite research skill for Claude?

Everything above depends on one file. The elite research skill is a document I have built and tweaked over roughly 6 months, and it is the difference between asking Claude to “check this research” and running an institutional-grade audit with repeatable standards. A skill, mechanically, is just a structured instruction file the model loads when the task calls for it. The leverage is in what the file encodes.

Mine has 6 sections. Here is what each one does and why it exists.

1. What sources should an AI research skill be allowed to cite?

The skill’s foundation is a closed list of approved sources, organized in 3 tiers. Tier 1 is elite publications: the New York Times, the Financial Times, the Wall Street Journal, the Economist, Harvard Business Review, MIT Technology Review, and MIT Sloan Management Review, each listed with its domain and its specific strength. Tier 2 comprises academic research centers: Harvard, Stanford, Princeton, Yale, Chicago Booth, and LSE, each with named search targets, such as specific working paper series. Tier 3 is institutional investment research: Goldman Sachs, Morgan Stanley, Bridgewater, Blackstone, AQR, BlackRock, JP Morgan, plus the top VC firms, a16z, Sequoia, Benchmark, Bessemer, and their peers, with a rule to pull published essays and market maps rather than portfolio announcements.

The operative word is exclusively. The skill instructs the model to cite nothing outside the architecture unless I explicitly ask. This single constraint kills the SEO-farm citation problem before it starts. Most AI research quality problems are source-selection problems wearing a disguise.

Every finding then carries 1 of 4 credibility tags: [PEER-REVIEWED] for Tier 2 academic work, [INSTITUTIONAL] for Tier 3 research, [EDITORIAL] for Tier 1 reporting, and [OPINION] for op-eds and commentary, which must still come from approved sources. I can scan a brief and weigh every claim by the tag beside it.

2. How does the skill match research depth to the task?

The skill defines 4 modes, each with trigger phrases so the model selects the right one from how I naturally talk, each with its own numbered method.

Thesis Research activates on phrases like “I’m writing about” or “I want to argue that.” Its method: restate the thesis in 1 sentence, find supporting evidence, then deliberately hunt complicating evidence, then historical parallels, then 2 to 3 quotable data points, then any disagreements between sources. The output emphasis is narrative ammunition for an article.

Market Intelligence activates on client names and phrases like “competitive landscape.” It works through recent Tier 1 coverage, institutional research, relevant academic work, then risks and opportunities not yet in mainstream coverage. Output emphasis: decision-ready intelligence for advisory work.

Trend Scanning activates on “what are the smartest people saying about” and similar. It restricts to the last 30 to 90 days, then maps consensus, divergence, and the narrative that is forming but has not crystallized. Output emphasis: compressed, newsletter-ready signal.

Deep Dive activates on “comprehensive research” or “everything on.” Broad search, findings organized by sub-theme, a timeline where the topic has one, the key debates mapped, the gaps named, and a meta-narrative synthesized at the end. This is the 10-plus-source mode, and it is the one the dot-com project ran in.

The modes matter because “research this” is not one task. It is at least 4 tasks with different depths, different recency requirements, and different outputs. Naming them lets the file route the work.

3. How should AI research handle search queries and paywalled sources?

The skill specifies exact search patterns, site: operators per source, working-paper queries for the academic tier, and firm-name patterns for institutional research. It sets a floor of 5 searches per request, more for Deep Dive. It mandates recency within 12 months unless the topic is historical.

It also contains the section I consider non-negotiable: paywall discipline. Most Tier 1 sources are paywalled. The skill instructs the model to attempt the full fetch, and when blocked, to extract only what is actually visible, provide the direct URL for me to open through my subscriptions, and tag the finding [PAYWALLED — link provided]. The controlling rule is on one line: never assume content is behind a paywall. Only report what you can actually see. Fabricated paraphrases of paywalled articles are among the most common and least detectable AI research failures. This section is the immune response.

4. What output formats should AI research produce?

Three fully specified templates, chosen by destination. Format A, the Research Brief, feeds articles: thesis, key findings with tags and full citations, complicating evidence, historical parallels, quotable data points, source disagreements, narrative hooks, and a list of paywalled sources to review. Format B, the Advisory Intelligence Memo, feeds client work: executive summary, market context, a key-intelligence table where every row pairs a finding with its source, credibility tag, and a “so what” implication, then competitive landscape, risks and blind spots, and recommended discussion points for the client call. Format C, the Signal Pack, feeds the newsletter: the signal in 1 paragraph, top insights at 2 sentences each, where the smart money disagrees, the emerging narrative, and a prioritized reading list.

Templates sound bureaucratic until you run without them. Unstructured research output is a wall of prose you excavate. Structured output is a component you install.

5. Why should AI research output never hedge?

The skill enforces the Operating voice on the research output itself. Declarative sentences, no hedging, active voice only, findings stated as facts. “Goldman’s research shows X,” never “Goldman’s research seems to suggest X.” A banned-word list that includes “perhaps,” “arguably,” and “it seems.” The reasoning: a research brief is not a neutral academic document. It is intelligence, written with an operator’s eye for what matters. Hedged findings defer the judgment work to future me, and future me is busy.

6. How do you quality-check AI research before delivering it?

The skill closes with 8 verification items the model must clear before delivering anything: every finding cites a specific approved source, every finding carries a tag, nothing off-architecture slipped in, paywalled items are flagged with URLs, the format matches the destination, the voice complies, source disagreements are surfaced rather than smoothed over, and the output is actionable rather than merely informational.

That 7th check deserves a sentence. Models are consensus machines. Left alone, they average conflicting sources into a smooth, confident, wrong middle. Forcing disagreements to the surface is where the most interesting angles live. Goldman versus Bridgewater is an article. Their average is nothing.

The skill also declares its integrations, it feeds my meeting-intelligence pipeline, hands off to my brand system for visuals, and supplies the source architecture when my contrarian filter needs counter-evidence. Skills compound when they reference each other. A lone skill is a tool. A connected set is a staff.

How do you build your own AI research employee?

You do not need 6 months to get a working version of this. With my Oper(AI)te co-instructor, Wessal Khader, I have built the whole construction process into a set of workbooks called Your Research Employee, attached to this post in 4 editions:

Each workbook is a file that an AI can run. You upload it, paste 1 kickoff prompt, and the AI walks you through the build as a guided interview, 1 question at a time. No fields to fill in by hand. The build takes about an hour across 4 steps:

The Research Standard, about 20 minutes: Nine questions that force specificity: the decisions your research actually feeds, the output shape you need, when a fast answer suffices versus a deep dig, which sources you trust on sight, which are noise to you, your domain, and whether you want a recommendation or evidence laid out for your own call. The test of a good standard is that it names a real decision. A brief with no decision behind it is a Wikipedia page.
The Source Rules, about 15 minutes: The side nobody teaches. The workbook ships with a measured floor built from published research on how AI research fails, including the Columbia Tow Center finding that 8 AI search engines got the source wrong more than 60 percent of the time across 1,600 queries. On top of that floor sit 9 banned research habits, uncited claims, single-sourcing, stale data stated as current, marketing repeated as fact, and 5 more, each paired with the positive rule that replaces it.
Sources and Assembly, about 15 minutes: Your trusted-source list, your internal documents, and your question workflow get tested and assembled into 1 paste-ready analyst profile: standard, rules, sources, mode, output format.
The Source Audit, about 10 minutes: A repeatable protocol for scoring any brief against a fixed table. 0 to 1 violations are decision-grade. 2 to 4 means fix the claims and update your rules tonight. 5 or more is a guess, wearing a citation. Each workbook closes with a swipe file of 6 prompts for daily operation, including my favorite, the test drive: make the analyst research a question you already know the answer to, then grade it against reality.

The 4 editions exist because the 4 platforms do this job differently. Claude is the only one that wires your private data in as sources through connectors, and it thinks the longest before answering. ChatGPT produces the strongest long-form synthesis and is the most confident when wrong. Gemini brings the biggest context window and the tightest Google Docs handoff. Perplexity tested most accurately on citations and holds a standing persona most loosely. No edition is best at everything. Pick the edition for the machine you live in, and if you run several, the steps are the same on each.

These are the files we will be using in Oper(AI)te this Summer, starting this Tuesday, the 21st of July

The Claude Edition: The only edition that connects your private data as sources through connectors.

The ChatGPT Edition: Built around Deep Research and Memory.

The Gemini Edition: Built around Deep Research with Google grounding and Workspace.

The Perplexity Edition: Built around Research mode, Labs, and Spaces.

Grab Your Seat Before They're Gone!

What should you do with this workflow this week?

The pipeline, compressed: commission research with a real brief that demands chartable data. Move the output to a second, different model and audit it against a written standard before you trust a number. Rebuild the visuals in your own identity. Produce the final document in your own voice, then read every word yourself before it ships.

This week, run 1 piece of it. Take any research you are currently relying on, drop it into a different model from the one that produced it, and ask for an audit against the workbook’s source rules. Count the violations. The number will tell you whether you have research or a guess with a citation.

And if you want to build the full system live, this is exactly what we teach in Oper(AI)te, the Summer of AI Fluency cohort I am running with Wessal Khader. Six live builds: a research analyst, a voice employee, a meetings employee, a visual brand, wired into the tools you already use. We start Tuesday, July 21. Through the deadline, code TAKE500 takes $500 off, $1,999 instead of $2,499. Details at unfazedfounder.com/operaite.

Grab Your Seat Before They're Gone!

The tools will keep improving without your permission. The standards will not.

Frequently Asked Questions

Can you trust AI research from Perplexity without verifying it?

No. Perplexity Computer is the strongest collection engine I use; every claim arrives attributed, and it still shipped 5 substantive errors in this single report, including a 22-point understatement of Cisco’s crash and a mislabeled valuation ladder for Anthropic. Attribution is not accuracy. Audit everything that feeds a decision or a publication.

Why should you verify AI research with a different model than the one that produced it?

Two models do not share the same blind spots. A model auditing its own output tends to re-approve its own errors, because the same weights that produced the mistake evaluate it. Collect in one place, then verify in another, ideally with a stronger model than did the collection.

Which AI is best for research: Claude, ChatGPT, Gemini, or Perplexity?

They do different jobs. Claude is the only one that reads your private data as sources through connectors, and it thinks the longest before answering. ChatGPT produces the strongest long-form synthesis and is the most confident when wrong. Gemini brings the largest context window and the cleanest Google Workspace handoff. Perplexity tested most accurately on citations. The staff you want runs each job on the machine that does it best.

What is a Claude skill?

A skill is a structured instruction file that the model loads when a task calls for it. Mine encodes an approved source list, 4 research modes, search and paywall rules, 3 output templates, voice rules, and 8 quality checks. The file replaces my memory and my mood as the quality-control layer.

What are the most common errors in AI research?

Wrong citations delivered confidently. The Columbia Tow Center found 8 AI search engines got the source wrong more than 60 percent of the time across 1,600 queries. After that: fabricated links, stale data stated in present tense, single-sourced claims, and vendor marketing repeated as fact. Every one of these is testable, which is why the workbooks ship with a source audit.

How long does this workflow take?

The machine research took 33 minutes. The full pipeline, audit, charts, and write-up included, fit in an afternoon, plus the time it takes you to read the final report in full, which is not optional. Building your own research employee from the attached workbooks takes about 1 hour.

Do you need to know how to code to build this?

No. Every piece of this system is a document and a set of prompts. You upload a workbook, paste 1 prompt, and answer questions in plain language. The most technical act in the entire pipeline is downloading a file from one tool and attaching it to another.

About John

John Brewton documents the history and future of operating companies at Operating by John Brewton. He is a graduate of Harvard University and began his career as a PhD student in economics at the University of Chicago. After selling his family’s B2B industrial distribution company in 2021, he has been helping business owners, founders, and investors optimize their operations ever since. He is a career consultant and business operator, asking the question: What is the future of companies?

Connect with John on LinkedIn

How To Stay Relevant In Our AI-First World w/ John Brewton & Wessal Khader

John Brewton — Wed, 15 Jul 2026 18:08:45 GMT

Available for iOS and Android

How to Actually Make AI Write Like You, Not Like a Robot, With the Accent of Your Favorite LLM.

John Brewton — Mon, 13 Jul 2026 14:03:21 GMT

The 25 Most Important AI Jobs I Run For Myself & Build For My Clients Every Week

John Brewton — Tue, 30 Jun 2026 13:32:16 GMT

TL → DR

Most people run 2 jobs with AI. They ask a question and close the tab.
I run 25. Each one is a skill you can own, not a prompt you retype.
The list below is the whole staff, sorted into 5 groups.
The gap between 2 and 25 is the difference between using AI and running it.
11 of the 25 run as autonomous loops, several agents working together with no ha…

AI Fluency, Defined: The 5 Moves That Replace Prompt Lists

John Brewton — Sat, 27 Jun 2026 11:32:21 GMT

TL → DR

AI fluency is the capacity to see the whole workflow and build it yourself. Not knowing a tool. Operating one.
Prompt lists are a phrasebook. They run out. Fluency is a grammar. It does not.
The discipline is 5 moves: Frame, Assign, Brief, Inspect, and Iterate then persist.
You drill them on small prompts until they are automatic. Then the same 5 moves run multi-agent systems.
Paid subscribers get the tool: the Rudiments skill for Claude and ChatGPT, plus a one-page chart.

Subscribe now

I live in a bilingual household. My wife is from Mexico; Spanish is our first language at home, and our two-and-a-half-year-old daughter is becoming bilingual day one. She moves between Spanish and English as if it’s all one language. She produces sentences no one taught her. It’s one of the most beautiful experiences of my life.

I studied Spanish for 4 years in high school and Chinese for 4 years in college. 8 years of classes, and I could not have held a conversation in either. I never became fluent. I was memorizing lists for the test and forgetting them by summer. Since meeting my wife, I have worked hard to become bilingual. I’m not quite there yet, but getting very close. What is making me fluent now is my household, my wife, our daughter, our friends, and my extended family. I speak more Spanish than English in my personal life each day.

That is how most people should learn AI, and almost no one does. They collect prompts instead. “10 prompts for marketers.” “The 50 best ChatGPT prompts.” They are building a phrasebook, and they will struggle to produce high quality outputs the first time the work is not on the list.

AI fluency is the capacity to see a workflow whole and build it yourself, which is the one thing software has spent your entire career doing for you.

Nobody has defined AI fluency, so the word floats. People feel behind without knowing what they are behind on. Companies buy licenses and wait for a result that never arrives.

The gap is not access to the tools.

The gap is a definition nobody wrote down or thought to calrify.

What does it actually mean to be fluent?

Linguists test it with 3 markers:

A fluent speaker stops translating in their head.
They route around the words they do not know instead of freezing.
They create instead of reciting.

My daughter is 2 and already does the third one. Most adults with a language app never clear the first.

The same 3 markers tell you a person is fluent with AI. They stop reaching for a saved prompt. They route around what the model cannot do instead of giving up. They compose a workflow for a problem no one has templated.

New to Operating? You Might Appreciate These Popular Articles

Why is working with AI different from using software?

For 40 years, enterprise software has made this skill unnecessary. The ERP told you which screen to open. The CRM told you what to enter and when. The workflow was built into the software, written by a vendor who had already anticipated your situation. You pointed at the phrase.

That arrangement ended in 1 step. The model does not hand you a directional form on a screen; it hands you a monumentally flexible canvas through which you can communicate to build anything, literally.

The cost of building a workflow fell to a sentence you can type. That changed what a tool is. Which changed what a worker does. This is now changing what it means to be a worker. Most people are still asking which prompt to use.

We have spent our careers doing work that computers could not do, yet. Now the computer does the work and asks us to compose and direct its intended output. That is not a smaller job. It is the work of an Operator, and very few companies, much less individuals, have been trained for this task.

Can you become AI fluent without a technical background?

I trained as a drummer at the Berklee College of Music for a year right after high school. I obsessively drilled rudiments, single strokes, double strokes, paradiddles, slowly and cleanly, thousands of times, until my hands moved without thought.

When the rock-and-roll dream didn’t work out, I became an equally obsessive student, ultimately graduating from Harvard with a degree in Economics. The disciplines and habits that helped me master an instrument were the same tools that helped me succeed in academia.

I did it again in business. I spent a decade at a B2B industrial distributor, from the sourcing desk to the chief executive role, and grew it from $27.5M to $75M before selling in 2021. The work computers could not do yet was the work I did by hand for 10 years.

I am doing it a third time now. I am building AI-first, enterprise-grade software using tools Attio, Claude, Supabase, and Resend. I am not an engineer. I am an Operator, a buildeer, a designer, and a founder.

The discipline that produced all 3 outcomes across my life transfers and compounds with each successive iteration.

For AI Fluency, this skill-building repetitive loop has a clear shape, I call it the Five Moves. It is rudimentary, and meant to be practiced over and over again

You do not become AI Fluent by reading about reps. You achieve this new reality by building this capability across your real work, alongside colleagues who are doing the same, and by learning from leaders who have already achieved the outcome.

That is what I built with Wessal Khader. It is called Oper(AI)te, and it runs this summer. 8 live sessions, no theory, and no replays to watch alone. You build a working staff of AI employees that does your real work, across Claude, ChatGPT, Gemini, and Perplexity. Beginner or advanced. No coding ability required. You leave with a team of agentic AI-employees running 24/7.

The summer cohort starts July 21, and seats are limited.

Take Your Seat @ Oper(AI)te!

Build it with us this summer to become AI-Fluent by the time Q3 ends and Q4 starts!

What are the 5 moves of AI fluency?

The five never change. Only the size of the task changes. You drill them on a one-line prompt the way a drummer drills a single-stroke roll, then the same 5 moves run a multi-agent system. Start small on purpose. A weak prompt costs nothing, and you can run the loop 40 times in a morning on real work.

Frame: One question: what is the actual outcome, and what does done look like? Small: “summarize this” becomes “pull the 3 decisions and who owns each.” Large: map where the workflow begins, ends, and every touchpoint between. Skip it and you accept the first plausible thing the model returns, because you never defined better.
Assign: One question: who does each part, you, the model, a tool, or an agent? Small: decide whether the model is thinking or just formatting a decision you already made. Large: the model’s job, a tool’s job, a sub-agent’s job, your judgment. Skip it and you hand the model work it cannot do, then blame it for guessing.
Brief: One question: what do you know that the model does not, and what must it not do? Small: paste the constraint, the audience, and one example of good. Large: persistent context, data sources, guardrails. Skip it and the answer comes back fluent, confident, and generic, because you described a generic situation.
Inspect: One question: where is this wrong, thin, or bluffing, measured against move one? Small: read the output as an operator checking work, not a customer accepting a delivery. Large: test points, spot checks, the places a workflow breaks quietly. Skip it and fluent-sounding wrongness ships. This is the most expensive failure in the discipline.
Iterate, then persist: One question: what one change improves this, and how do you keep what works? Small: change one variable, run again, notice what changed. Large: save what worked as a prompt, a project, a skill, or an agent, so it runs again without you. Skip it and every task starts from zero, and nothing compounds.

How do you practice AI fluency?

The cost of a mistake on a 2-line prompt is limited. The rep count is everything. Run the loop 40 times a day on real work, and the moves leave your conscious attention. By the time the task is a system with sub-agents across 4 tools, the moves are already in your hands, and your attention is free for the one thing that does not automate: judgment.

The rudiments are rudimentary. That is the point. Mastered, they become the foundation for building things that run without you.

Is there a tool that enables the 5 moves?

I built a skill that runs you through the 5 moves while you work. It is called Rudiments. When you sit down and say “help me draft the board update” or “build me a lead-routing flow,” it does not lurch into the task. It asks you the moves, calibrated to the size of the job, and it shows you the difference between your loose ask and the sharp one, so you learn the move and not just the answer. It is built to make itself unnecessary.

Here is what it does with every task you bring it:

Sizes the job. A quick rep for a one-line prompt, a full rep for real work, a build rep for a multi-agent system.
Walks you through the 5 moves as questions: Frame, Assign, Brief, Inspect, and Iterate then persist.
Shows you your loose ask next to the sharp one, so the rep teaches you the move.
Tracks the moves you skip and tells you which one to drill next.
Saves what works, so the rep compounds instead of resetting to zero.

That is the loop and the rep that will help you get one step closer to becoming AI Fluent.

You do not need to write code. You do not need a developer. You need Claude Pro at $20 a month or a Premium ChatGPT account, and 60 minutes of real work to drill against.

Paid subscribers get 3 things, attached below:

The Rudiments skill for Claude, installed in under a minute.
The same discipline written as instructions for a ChatGPT, so it runs wherever you work.
The one-page chart, print-ready.

Become a paid subscriber to download the Claude skill and the directions for how to build the same capability in ChatGPT after the frequently asked questions section below.

Subscribe now

Frequently Asked Questions

What is AI fluency? AI fluency is the capacity to see the whole workflow and build it yourself. It is not knowing a tool. It is operating one. The test is the same as for language. You stop reaching for a saved prompt, you route around what the model cannot do, and you compose a workflow for a problem no one templated.

What is the difference between AI fluency and prompt engineering? Prompt engineering collects lines that work. AI fluency builds the workflow that those lines sit inside. A prompt list is a phrasebook, and it runs out. The 5 moves are grammatical, and they do not.

Do you need to know how to code to become AI fluent? No. I am building an AI-native CRM for a public company, and no one trained me to write software. You compose the workflow in plain language and let the model and tools handle their parts. You need judgment, not a computer science degree.

What are the 5 moves? Frame, Assign, Brief, Inspect, and Iterate then persist. Frame the outcome. Assign each part to you, the model, a tool, or an agent. Brief the context and the guardrails. Inspect the output against the outcome. Iterate one change, then save what works so it runs again without you.

How long does it take to become AI fluent? Faster than you expect, because you drill on small tasks with a high rep count. Run the 5 moves 40 times in a morning on real work, and the moves start to leave your conscious attention in weeks, not years.

Should I use Claude or ChatGPT to practice? Either. The discipline is identical on both. Claude Pro is $20 a month, and ChatGPT has a free tier. The Rudiments skill ships for Claude, and the same discipline is written as instructions for a ChatGPT Custom GPT.

What is the Rudiments skill? It is a tool that runs you through the 5 moves while you work, calibrated to the size of the task, and shows you the difference between your loose ask and the sharp one so you learn the move. It is built to make itself unnecessary. Paid subscribers get it for Claude and ChatGPT, plus the one-page chart.

Download the first skill you need to become AI Fluent below.

How to Build Your Own AI Employees: A Simple 2026 Guide

John Brewton — Sun, 21 Jun 2026 00:17:56 GMT

What is this article about? (TL → DR)

Most people use AI like a search box. They ask one thing, then close the tab.
A few people do more. They build AI “employees” that do the same work every day.
An AI employee is a setup you create once. It learns your voice, your job, and your rules.
Four are worth building first: one for writing, one for research, one for meetings, and one for your own job.
You can build the first one this week. No code. The free kits and a free live class are at the bottom.

Want to build one live with me, for free?

Register for my free Masterclass on June 16th, where I will teach you how to build your own AI Employee that runs your calendar and inbox! Register Here

A free 60-minute live class with me and . You come in with a full calendar and a packed inbox. You leave with an AI employee that runs both, and the way to build the next 10. No code. No replay.

Reserve Your Free Seat

What is the difference between using AI and operating it?

Using AI is simple. You open it, ask a question, and close it. The answer sounds flat. The tool knows nothing about you.

Operating AI is different. You build something that stays. You teach it your voice, your job, and your rules. Then you run it like a teammate who already knows your standards.

Here is a typical Monday for a person who built their staff:

A big slide deck is due on Wednesday. It gets drafted in an hour, not over two nights. Three calls run back-to-back.
Each one ends with the notes, the choices, and the next steps already written.
A question that used to eat up an entire afternoon comes back with a sourced answer before lunch.

That is not a dream. That is what the right setup does.

What is an AI employee?

An AI employee is not a prompt you copy from Reddit or copy and paste from a LinkedIn post. It is a setup built for one job inside your team and workflows. You engineer it one time. It runs the same way every time. You adjust and tweak, but do not ever need to rebuild.

These are the four most worth building first:

The Voice Employee writes like you. You train it on your real emails and posts. It can write to a client, to your team, or to your boss, each in the right tone. It checks its own work before you see it. Done right, no one can tell you used it.
The Research Employee gets you ready. You give it the questions you ask every week. It pulls real facts, sorts them, and hands you a short report with sources. It preps you before any meeting that matters.
The Meetings Employee kills the blank slide. You give it your rough notes. It hands back a clean deck you would actually send. It writes the agenda. It turns your call notes into a summary. Every meeting ends with the work already done.
The Role-Specific Employee is yours alone. You have one task that eats hours it should not. A report. A proposal. A weekly grind. No template fits it. You build this one around your real job.

Connect With Me On LinkedIn

Connect on LinkedIn

Why do most people get AI wrong?

Most people stay stuck for one reason. They think of AI as a tool. Tools are passive. You pick them up, use them, and put them down.

An employee is different. It carries what it knows. It knows your standards. It knows what you would say no to. You do not explain yourself again every time.

Change that one idea, from tool to employee, and everything changes. How you build it. How you talk to it. What you expect to receive back.

Consulting Perspective: This is one of the biggest mental hurdles I see inside my clients’ companies. Individuals struggle to provide enough direction, detail, and specificity when they’re prompting one of the models to perform a taks, or within the build process for the engineering of their agent.
When you make the jump to looking at the AI as a new hire, a talented, smart analyst, and begin providing said analyst with all of the context, reference materials, and clear direction that you would to any team member you were asking to produce high-quality work product, that is when the quality of the results, the outcomes, and the strength of the builds you are engineering, dramatically improve.

The primary apps are not hard to use. Claude is strong in long writing and voice. Perplexity is built for research with sources. ChatGPT does a little of everything very well and is currently the cat’s meow for image generation. Gemini sits within Google, where many people already work, and has a unique capability to help with management and team-dynamics queries. You do not need to master one. You need to know which one to grab, and why.

Are most people actually good at AI?

Here is the hard part. Most people in your meetings are faking it. They put “AI” in their slides. They say they are “trying the tools.” They paste prompts into free accounts and take screenshots of the results.

The gap is getting larger. You can see it in who writes the clearest report the fastest. Who shows up to the meeting with the work done? Who gets more done without longer hours? That person, most likely, quietly built their staff.

The most costly move right now is to wait until it is obvious. By the time the gap is loud, it has already cost you.

How do I start building my AI staff?

You can build your staff this summer. Do it before the calendar fills and the year resets. Do it before a skill gap turns into a career gap.

Start with the Voice Employee. Train it on 10 to 15 real things you have written. Tell it what you never say. Tell it how you sound when you are rushed. Tell it what “clean but not stiff” means for you. Run 30 drafts before passing judgment.

What are the AI staff kits?

I built the four employees as kits so you do not have to start from scratch. One kit per employee: Voice, Research, Meetings, and Role-Specific.

A kit is a skill file. It holds the steps that run the job, the notes that carry your standards, and a template for the output. That is the gap between a prompt you retype and an employee who already knows the work.

Each kit has 2 modes. BUILD sets up your profile the first time. Your profile is the file that holds what the kit learned about you. OPERATE runs the job at your standard after that, and checks its own work before you see it.

How do I install and run the kits?

First, install it. If your app takes skill files, load the kit. If your app uses projects instead, paste the steps into a project and add the notes as files.

Second, run BUILD one time. Answer the questions. For Voice, that means 10 to 15 writing samples. For Research, your weekly questions. For Meetings, your formats. For Role-Specific, the one task worth it. This makes your profile. Keep it. It is the part that matters most.

Third, run OPERATE from then on. Give it the input. Take the output. Fix what missed. Save the fix back into the profile. It gets faster each time.

Start with the Voice Employee. Build it this week. Run 30 drafts before you judge it. Then add the next one. Each one makes the next build faster, because the work you did once carries over.

That is the whole setup.

Why does my AI output still sound generic?

The build misses in 2 ways. Both are quick to fix.

The skill does not run. The app only uses a skill when the description tells it when to. If nothing happens, just say it: “Use my Voice Employee to draft this.” Then check two switches. The skill and code tools are turned on in settings. One of those is almost always the cause.

The output still sounds robotic. That is a thin profile, not a broken tool. Add 5 more writing samples. Use the ones you wrote in a hurry. Make the “never say” list longer. Name 3 phrases you would never use. Run 10 more drafts. It gets sharp from your fixes, not from the first try.

When the output misses, fix the profile, not the prompt. The prompt stays the same. The profile is the part that learns.

Want me to build it with you?

We take one problem in your business and finish it across four sessions. You pick the problem: your pricing, your brand, your content plan, or one issue inside a company you already run. We name it, we plan it, we build the machine inside Claude or ChatGPT, and we put it to work on real tasks.

It normally sells at $999. The first 10 seats are $499, half off. Each seat includes a full year in the Operating Founders community in Skool and weekly Friday office hours. You can run your four sessions any time through December 31, 2026.

Claim Your Seat

Frequently asked questions

Do I need to know how to code?

No. You do not write code. You answer questions and paste in samples. The kit does the rest.

Which AI app should I use?

Any of the big ones works. Claude is great for writing and voice. Perplexity is great for research. ChatGPT and Gemini both work too. Pick one and start.

How long does it take to build the first one?

About an hour to set up the Voice Employee. Then run 30 drafts to make it sharp. You can do it in a week of short sessions.

Is this safe with my private writing?

Your profile lives in your own account. Only load kits from people you trust. Read the files before you turn them on.

What if the output still sounds robotic?

Add more writing samples and a longer “never say” list. Fix the profile, not the prompt. It learns from your fixes.

How much does it cost?

The kits are free. A paid AI plan runs about $20 a month. A free plan is enough to start.

Do I have to build all four at once?

No. Build the Voice Employee first. Add the others when you are ready. Each one makes the next faster.

What is the difference between a prompt and a kit?

A prompt is one message you retype. A kit is a saved setup that already knows your job and runs the same way every time.

Where do I get the kits?

Have a look at the Field Guide here and then become a paid subscriber to access all of the skill kits and files below.

Become a paid subscriber for as little as $7.99 per month and access them below…

Subscribe now

The 17 Questions That Tell You Whether You Are AI Fluent

John Brewton — Tue, 02 Jun 2026 11:37:11 GMT

TL · DR (The 30-second version)

AI fluency is not knowing what AI can do. It is changing how your work actually gets done. Knowing the capability exists changes nothing on its own.
The research splits in two. Stanford finds 14 to 34 percent task-level gains. NBER finds 89 percent of executives see no enterprise boost. McKinsey finds only 6 percent capture…

The Three Practical Builds To Become AI Fluent

John Brewton — Fri, 29 May 2026 15:07:10 GMT

Thank you , , , and many others for tuning into my live video! Join me for my next live video in the app.

Available for iOS and Android

Build Your Team of Agents

John Brewton — Fri, 29 May 2026 11:46:02 GMT