Entity of One: Monopoly Thinking for Experts in the AI Era

Pick a niche, claim a lane, call yourself the number one. The advice is everywhere. The resistance is stronger than anyone admits. Monopoly thinking sounds like a performance: chest out, voice loud, stake planted. That is not a personality most serious experts want. So they stay broad, stay generic, and lose the compounding that only goes to the few who dared narrow down. The premise is wrong. Monopoly thinking is not about ego at all. It is about being the only obvious choice for a tightly defined promise. In 2026 it is structurally easier than ever, because the systems that now answer most professional questions are entity engines. The expert who becomes the clearest entity for a promise is quietly handed the category while the loudest competitor is still shouting. We need a name for what that actually looks like. Call it Entity of One.

Three books define the classic position on monopoly thinking for experts. Peter Thiel's Zero to One (2014) is plain: 'Every startup should start with a very small market. It is easier to dominate a niche market than a larger, pre-existing one.' The sequence is clean: start small, dominate, expand, entrench, keep innovating with monopoly profits. Daniel Priestley's Key Person of Influence goes further and uses niche as a verb: 'niche then pivot, niche then pivot.' Not 'financial planner' but 'financial planning for rural family farms.' Joe Calloway's Category of One (2003) framed it differently again: escape commodity pricing by becoming an entire category, the place where comparison shopping stops making sense. Every serious expert should read all three. The principle underneath is sound and has aged well. Start with a market you can actually dominate. Build an unmistakable reason someone would hire you and no competitor. Escape competition by becoming the only obvious choice for a tightly defined promise. But all three books predate the way humans find experts today. In 2014, when Zero to One appeared, you were found through Google's blue links. Ranking for a keyword was the battlefield. The word niche fit that world: a slice of keyword space you could dominate because nobody else had bothered to write the ten pages. That is what the SEO era absorbed and flattened the advice into. A dozen years of keyword-volume tools, long-tail strategies and rank-tracker retainers turned 'niche' into a defensive word. It began to evoke a hunched corner of the market, not a durable monopoly. In 2026 the battlefield is different. Most professional questions are first answered by a language model. The model does not rank keywords; it clusters entities. Your presence is no longer a position on a results page; it is a node in a web of associations the model has learned from your name, your work, and the company you keep. A niche in the 2012 sense still exists, but it is a pale shadow of the real move. That real move is to become a recognised entity with a consistent promise. Entity is the right word in 2026 not because it is jargon but because it is what the models actually read. A language model building an answer in your field is asking: who are the entities associated with this question, who co-occurs with them in trusted sources, whose unique claims keep turning up. If you are a clear entity with a clear promise, you are picked up. If you are a smear of keywords, you are filtered out. The classics were right about the principle. The vocabulary is the only thing that needs an upgrade.

An Entity of One is the status of being the clearest, most internally consistent entity for your promise, in both human memory and the models that broker most of today's professional trust. It is monopoly, Category of One, Key Person of Influence, updated for an era where entities are clustered, not keywords ranked. The framework has three layers, stacked. Layer one: you, the person-entity. Your name, your face, your voice, your known points of view, the way you write and present, the questions you answer and the ones you refuse. The human node. Layer two: your business-entity. The company, product, or service that carries your work into the world. The offer, the pricing, the delivery, the way the brand shows up across channels. Layer three: the promise-entity. The one sentence that explains what transformation you deliver. Not what you do. What changes about the customer because of you. Each layer on its own is unremarkable. Plenty of people share your name. Plenty of businesses share your field. Plenty of promises sound like yours in isolation. The specific stack of all three is what becomes unmistakable. When the three layers are tightly aligned, you show up in trained models, in retrieval, in real search, and in human memory as one coherent thing. When they are scattered, you show up as noise.

Junior Veiga Gomes runs Twenty6 Suits in the Netherlands. Bespoke suits and smart casual for men, starting around thirteen hundred euros, appointment only. The market is ancient and crowded. Every Dutch city has a tailor with thirty years of experience and a polished wooden floor. His promise is one sentence: I dress superheroes. The boutique writes it this way: 'Je komt binnen als een gewone man en verlaat onze boutique als een superheld.' You walk in as an ordinary man and leave the boutique as a superhero. Not a suit. A transformation. Run that sentence through the three layers and it becomes immediately clear why it works. The person-entity is Junior Veiga Gomes, a specific face and tone carrying his own name. The business-entity is Twenty6 Suits, a specific shop with an aesthetic, a price point, and an appointment-only threshold. The promise-entity is not 'the best suits in Amsterdam' or 'bespoke tailoring for the discerning gentleman,' but 'from ordinary man to superhero, through the way you dress.' The detail most experts miss is where the narrowing sits. It is not in the target audience. 'Superhero' is not a narrow demographic. Almost any man with a reason to look sharp can project himself onto it. The narrowing is in the entity combination itself. There is no other bespoke tailor anywhere where precisely this person, this business, and this promise combine. The stack is the moat. In the 2012 keyword world this was awful positioning. 'From ordinary man to superhero' ranks for nothing. In 2026 the same sentence is a monopoly, because language models answering 'where do I buy a suit that changes how I feel' already have the cluster of associations they need, long before the keyword battle even begins.

The practical work breaks into five moves. Any expert who wants to be an Entity of One works on these for as long as they stay in their field. Move one. Write down what you say that others in your field do not. Not a niche. A specific, recognisable, slightly dangerous position that is unmistakably yours. If it does not make at least one colleague disagree, it is not specific enough. Move two. Work out what you can repeat for ten years without getting bored. Repetition is entity-building. An Entity of One says the same thing in the same way across years until the association fuses. The test is 'what will I still be saying when I am tired,' not 'what is smart to say.' Move three. Price like you know the answer. High-end, premium, never apologetic. Price is a signal the models and the humans both read. An Entity of One does not discount. Discounts blur the promise. Move four. Never compare, always describe. The telling move of an Entity of One is how they talk about competitors. They do not. They describe what they do, how they do it, what it produces. Comparison is the language of a commodity; description is the language of a category. Move five. Build frameworks around your intellectual property. Named frameworks, named stages, named principles. Give your ideas names and put those names on the same surfaces your own name appears. None of the five on its own produces an Entity of One. Together, across years, they do.

Twelve years ago, the SEO era punished breadth. Ranking algorithms rewarded depth in a small slice. Monopoly meant hyper-targeting. In 2026, language models reward a different shape. They reward clarity across an entity stack. They are willing to consider a wider promise as long as the entity behind it is unmistakable. This is why 'I dress superheroes' can work as a promise that is both broad and monopoly-strength. A wider promise, held together by a sharp person-entity and business-entity, produces a tighter cluster than any keyword niche ever did. The work is still hard. A real Entity of One takes years of consistent publishing, interviewing, and standing behind your work. But the reward now compounds at a different rate. Every podcast, every named framework, every interview, every post, adds another associative link to the cluster the models are building about you. Once the cluster is dense enough, the models do the heavy lifting of positioning you in every answer about your promise. That is the 2026 dividend. If you are an expert in 2026, stop hunting for your niche. Start building your Entity of One. Write down your three layers: the person, the business, the promise. Check whether all three align or whether they drift. Pick the five moves you will work on this quarter. Strip the word niche out of your marketing copy and replace it with a clearer description of your actual entity. Name one framework this month, and use it three times. The experts who keep hunting for niches in 2026 are running the 2012 playbook on 2026 infrastructure. The experts who build their Entity of One are running the 2026 playbook and will pull further ahead every year the models get better. You do not need a bigger ego to do this. You need a clearer entity.