Dan Koe’s latest essay makes a blunt case: if people do not know who you are or what you offer, there is little reason for them to pay you. In his view, the ability to capture attention and turn it into customers now sits above product polish, years of effort, or even professional self-improvement when the goal is to make more money.
He argues that this applies both to your own business and to work done for others. Companies are willing to pay heavily for that skill.
Attention as the core commercial skill
Koe frames the current environment around three choices people face as job security weakens: complain and do nothing, learn new skills and pursue a better job, or start building the business they have wanted to build. He believes the third group will grow sharply.
That, he says, is why so many people on social media talk about business through the lens of personal branding. Attention is finite, and in his telling it is getting scarcer. Money can be earned again. Attention that has already been spent cannot be recovered.
His conclusion is direct. Mastering attention on social platforms can support your own business, or it can become a service you sell as a one-person creative team.
Why he thinks the market is moving this way
Koe points to several data sets to show why he sees a large opportunity.
- Goldman Sachs projects the creator economy will expand from $250 billion in 2023 to $480 billion in 2027.
- In 2025, creators are expected to generate about $185 billion in advertising revenue, surpassing traditional media for the first time.
- U.S. brands spent $13.9 billion on creators in 2021, and the Interactive Advertising Bureau (IAB) expects that figure to reach $43.9 billion in 2026.
He writes from several angles at once: as a creator, as someone who has worked for creators, and as someone who has paid creators. Based on that experience, he says pricing in the market has become extreme. A 60-second spoken placement in a Reels clip or YouTube video can cost thousands of dollars, even though he says it often fails to produce results.
His broader point is that attention has shifted to social media. Radio, television, and news outlets no longer hold people in the same way. Whether the goal is a small business built around a personal interest or the next unicorn startup, publishing content has become a standard way to attract paying audiences.
He uses Skool as an example. Koe says he regularly sees Hormozi ads and that more than $1 million a month is being paid to creators to promote Skool. He adds that Hormozi is also a creator himself, which he says has materially helped the company’s growth.
He also points to the rise of “founder-led marketing” in startup circles. Stripped down, his argument is that one of the strongest current growth methods for startups is still the same thing: build a personal brand and publish content consistently.
At the same time, he warns against confusing isolated reach with durable attention. A single tweet or Reels post that gets 1 million views while everything else gets 1,000 views does not mean a brand has been built. It may show that the content worked once. It does not prove people care about the person or the broader body of work.
Instead of chasing one-off hits, he argues for building a content world that people want to keep exploring.
The rise of the one-person creative team
Koe says the content business is going through a major shift. He identifies himself as the founder of Eden, a tool used by creators and digital businesses to research high-performing topics, plan and schedule content, sell digital products, and run email lists.
He says the point of mentioning Eden is not to pitch it, but to explain the kind of data his team sees. Across billions of posts, they can observe which topics, openings, and content styles work.
According to Koe, freelancers and agencies also use Eden to run personal brands for clients, covering content production, email marketing, digital product launches, and sales funnels. In conversations with content agencies and media teams, he says several themes came up repeatedly:
- Teams have not necessarily started layoffs, but they can take on more work without hiring because output per person has increased.
- Creators and founders hire outside help because they do not have time to think through content while also running teams and companies.
- AI can help write, but clients still need a strategist or creative director who knows how to use AI better than they do.
- AI still does not write the best posts, in his view, but it can study what performs, understand strategy, and draft something close enough that the human role shifts toward editing and revision.
From there, Koe says the most profitable AI business for one person tends to take one of two forms.
- Be your own creative team: build a personal brand, grow an audience, sell digital products, and build an email list.
- Be someone else’s creative team: help companies, founders, and creators who do not have the creative bandwidth to produce distinctive content on their own.
He describes this as becoming the creative director for your own brand or for someone else’s. If the operator is skilled and willing to practice, he says serving one to three clients and earning $10,000 to $75,000 a month is not out of reach. In his framing, that fee can still look small compared with the cost of a full team.
The three roles inside that model
1. Brand strategist
This role handles positioning and outside perception. The job is to stop a client from chasing every trend or treating every post as a shot at virality. Growth and engagement matter, Koe says, but not at the cost of trust and authority.
2. Content strategist
This is separate from brand strategy. The content strategist studies what works across topics and niches, then recommends what the creator should publish next. That includes post formats, YouTube thumbnail patterns, title structures, completion rates, and shareability.
Koe says people in this role are often strong at audience growth. But without brand strategy, a creator can stop posting and few people will ask where they went.
3. Writer / social media operator
This person drafts tweets, YouTube or Reels scripts, long-form X posts, and Substack newsletters, then plans and publishes them. The work has to align with both the topic direction from the content strategist and the positioning set by the brand strategist.
For Koe, this is where many people misuse AI. They treat it as a shortcut to fast money, ask Claude to generate scripts without direction or data, and expect that output to connect with an audience. He says very few people think seriously about brand strategy. Some think about content strategy. Most simply turn scattered ideas into posts and hope they take off.
How he would onboard a client
The third section of the essay turns practical. Koe walks through how he would help a client build what he calls an irreplaceable personal brand. He presents the process as useful both for creators doing it themselves and for operators selling services such as ghostwriting tweets or writing Substack posts for brands.
Step 1: Run an onboarding call and define the client’s “thought signature”
To make a client’s brand and content stand out, Koe says you need a large amount of background information. After signing the client, he recommends a call to ask for everything needed, record it, and keep the transcript so AI can reference it later.
He says he has already prepared a quiz or prompt set for people doing this for themselves, along with a 30-day creator sprint course. The goal is to combine what is currently working in the market with what is unique about the person.
He then lays out a long list of questions that shape the brand and can generate enough material for six to 12 months of content:
- Who do you help? Describe a real person, their bad day, what they complain about, and what they have already tried that did not work.
- Your story: what have you gone through that your audience is going through now, what was the turning point, what have you succeeded at, failed at, tried, or figured out, and what results can you point to.
- Who shaped your thinking: which books, posts, or videos you never miss, which ideas stay with you, and who changed your thinking even if you no longer agree with them.
- Content pillars and unique angles: what topics you can discuss for hours, which two or three matter most, how your angle differs from others, what advice people ask you for, and what you repeatedly explain.
- Mission and beliefs: what you want to help people achieve, what you believe that most people in your field do not, and which common advice in the field you think is wrong.
- What you stand against: which lifestyle is hurting the people you want to help, what you want them to escape, and which myth or habit you would eliminate if you could.
- Why now: what has just changed in the world that most people have not noticed yet.
- Where you are taking them: what their daily life looks like after following you for five years, how they would describe themselves, and what role you play in their story.
- Rules and language: what your ideal reader believes and does every day, and whether you have named your ideas, methods, or recurring problems.
He suggests storing all of this in a safe place, such as a brand board inside Eden, so it can be referenced whenever new posts are being developed.
Step 2: If the client already publishes, collect their best content
Koe says AI cannot truly copy a person’s voice. Even when it gets close, something still feels off. For that reason, interviewing the client and collecting their best work is not about full automation. It is about building a knowledge base that contains what is distinctive and worth preserving about that person.
Once that exists, AI becomes much more useful for brainstorming, drafting, and ideation.
His suggested workflow is to connect Eden to Claude or ChatGPT through MCP so the model can pull a client’s posts or other creators’ high-performing posts. Then provide the client’s social accounts and use a prompt that asks the model to pull the best content from each platform, take 10 to 20 examples per platform, identify the core ideas that make the person different, note the distinctive parts of their voice, and infer the worldview or philosophy that could rally others around them. The output should be organized into a document.
He also says it is worth manually reviewing the client’s top-performing posts on each platform to find repeatable patterns.
Step 2.5: If you have not started publishing yet
For beginners, Koe says this may be one of the best things they can do. His reasoning is that a person’s thinking is shaped by the people who influenced them most. The ideas that drive growth often come from how someone interprets what they absorbed from others.
He compares that process to “stealing like an artist.” In the essay, he says much of his own thinking is a blend of ideas from Naval, Leo Gura, Mihaly C, Steven Kotler, and Ken Wilber, mixed together and then filtered through his own life experience.
So if someone has not started publishing, he recommends making a list of the five to 10 people who have influenced them most. If those people are creators, even better. Put them into a list so their content can be reviewed in one place or engaged with directly. Then use the same prompt structure with those accounts. If they are not active on social media, ask AI to extract their best ideas from the web and from their books, then save that material as a document.
Step 3: Build a data-backed content map
By this point, Koe says, two things should already exist: enough brand background to make something distinctive, and a record of the best ideas that can be reused, repositioned, or adapted. The next move is to build a content map supported by data.
He says many creative people dislike studying viral posts, top-performing YouTube titles, or content directions that are more likely to work. But he rejects the idea that this somehow cheapens the process. In his view, it is simply reference gathering. Artists do this all the time. They use inspiration boards and references, not to copy line by line, but to extract principles that improve the odds of success.
Koe again mentions Eden here, saying plainly that he does not think another tool does the same job. Standard social media scrapers do not show which content performs well across multiple accounts, and tools like GrokBot are limited to X. If the goal is to find distinctive topics, he says it helps to study platforms where you are not already publishing.
The prompt structure he gives is detailed: ask the model to create a data-backed content map with three to four broad but distinctive content pillars; under each pillar, five to seven topics; under each topic, five to 10 ideas likely to perform well; and under each idea, four to five angles. The model should read the brand document and content document, extract possible pillars and the creator’s perspective, then use Eden to research what has worked across platforms over the past six to 12 months. The final result should be an easy-to-read document.
He says that process can produce roughly six months of content ideas. From there, Eve or another AI tool can draft material using the background already assembled, while post structure can be checked against patterns from content that has already performed well.
His view is that it would be difficult for this process to produce no results at all, because it creates a large gap between the operator using it and the people who are not.
How to get the first client
Koe closes with an example from a few years ago. His friend JK Molina, whom he describes as a major force behind the rise of Twitter ghostwriting, sent him a direct message that said, in essence: “Dan, I know you do not want someone else writing for you, but I wrote 10 posts for you anyway. Try publishing them. If you like them, let me write more, or at least give me a testimonial.”
Koe says that before working in social media, he had been a freelancer and hated cold outreach. But that message changed how he thought about client acquisition. If the goal is to win a content client worth $2,500 to $5,000 a month, he says this is one of the best strategies available.
He suggests targeting creators, founders, or people whose growth has stalled. Then use a prompt to run through the full process described earlier, draft 10 posts for them, polish the work, and send 10 direct messages a day.
In his words, if someone actually does that, it is hard to believe they would fail to land even one client within a month.
He calls this the “Permissionless Apprentice” approach: do free work for your dream client, then find a way to put that work in front of them.
If direct messages do not work, he says the work can be made public. He gives an example of a Reels post that opened with: “Grant Cardone needs a new creative director. This is my job application. His content looks like it is still stuck in 2016.” According to Koe, Grant Cardone replied, “Come find me, I am stuck,” and the engagement on that post, in his view, supported much of the broader argument in the essay.
He ends by saying the opportunity is large right now. The tools needed to win clients, onboard them, and produce strong content for them are already available.
