Do not index
What should our AI content strategy be?
If the answer comes from the same vendors your competitors bought from, delivered by the same consulting firms running the same four-step framework, you do not have a strategy. You have the market consensus, invoiced as if it were bespoke. Companies are not just starting to sound the same. They are converging on the same decisions, and the sameness arrives upstream of anyone who would have had to approve it.
State of Brand made the distinction that matters, and it is worth keeping the two failures apart. Flattening happens to your words. A model compresses your category into a table, every vendor description reads alike, and you become interchangeable in the summary before a buyer ever reaches your site. Convergence happens to your decisions. It shows up as a deliverable that looks custom, costs like custom, and is the fourth copy of a document the same firm sold three of your competitors this quarter. The line from the piece: "Nobody decides to sound like everyone else. It happens upstream of whoever would have to make that decision."
The scale makes it concrete. Deloitte put Claude in front of 470,000 people. Accenture signed with OpenAI and then with Anthropic eight days later, and rolled Copilot out to around 743,000 staff. PwC ran a 200,000-seat ChatGPT rollout. Every one of those was announced as an edge over the firm next door. Read them together and none of them is an edge. They are the same purchase, made simultaneously, described as differentiation by four organizations that will now advise their clients using the same tooling.
Why this hits small operators harder than it hits Accenture
Deloitte can afford to sound like Accenture. It has distribution, capital, procurement relationships and decades of institutional trust absorbing the cost of sameness. You do not. If you run a business between $200k and $2M in revenue, sameness is not a brand inconvenience. It is the destruction of your only structural advantage, which is the ability to say something a large competitor will not say.
Here is the sequence I keep watching. A three person agency hires the same kind of AI help everyone else hired. It receives positioning that is defensible, professional, and identical to the positioning of eleven other firms in its category. Having nothing left to distinguish on, it competes on budget. That is the one fight a small business categorically cannot win, and it entered that fight by outsourcing the single decision it should never have outsourced.
This does not apply if your advantage genuinely is scale and price. If you win on operational cost and volume, converging on the standard toolchain is rational and you should do it faster than the firm next door. Skip this entirely if you are still looking for the right AI tool. The tool is not the variable. Everyone has the same tools within a quarter of each other, which is exactly the mechanism producing the problem.
The Only-You Test
The filter I apply, and I call it the Only-You Test, is a single question asked of any positioning statement, any content pillar, any strategy document. Could a competitor sign their name at the bottom of this and have it still be true?
If yes, it is not strategy. It is category description. Most of what gets sold as AI strategy fails this test, because a model producing it has only ever seen what already exists, and a consulting framework producing it is optimized to be defensible across many clients rather than correct for one. Both are structurally biased toward the middle of the distribution.
What survives the test is always the same category of material. The specific thing you did that nobody else did. The client outcome only you have the numbers for. The position you hold that costs you deals. The opinion you would defend in a room where it is unpopular. None of that comes out of a tool, because none of it exists in the training data and none of it exists in the framework. It exists in the founder's head, and getting it out is a different kind of work than buying a capability.
That is also why practitioner positioning outperforms thought leadership positioning so reliably right now. A practitioner has proprietary inputs by definition, which is the whole argument for positioning founders as practitioners rather than as thought leaders. A thought leader repackages what is already public, and repackaging what is already public is the thing that just got commoditized to zero.
The strategic implication is uncomfortable but clean. Everything a model can produce about your business is, by construction, everything the market already believes about your category. Buy that and you have bought the consensus. The value of proprietary input, the material that only exists inside your operation, rises every quarter as the cost of consensus output falls. Businesses that spend the next two years building systems to extract and publish what only they know will be the ones with pricing power. Businesses that spend it buying the same brain everyone else bought will find their differentiation was a line item that expired with the contract.
