Your next client may never type your name into a search box. They may instead ask an assistant a plain question — “can you recommend an executive coach for a first-time founder?” — and receive, in seconds, two or three names with a short reason for each. Understanding how that answer gets built is no longer a technical curiosity. It’s the newest room your reputation has to walk into.
The intern, not the oracle
It helps to drop the mystique. An answer engine behaves less like an all-knowing oracle and more like a very fast, very well-read intern under a tight deadline. It gathers pages relevant to the question, skims them, and drafts a confident-sounding answer from what it could actually absorb in the time available. Anything that slows the skim — clutter, ambiguity, thin content — gets set aside, not maliciously, just practically.
Three qualities make a page easy for that intern to use.
Retrievability
The page has to be fetched and parsed quickly, without megabytes of scripts or an interminable cookie negotiation first. Speed here isn’t a nicety; it’s a precondition for being read at all within the system’s time budget.
Structure
A real heading hierarchy lets the machine separate your credentials from your philosophy from your footer instantly — the digital equivalent of a well-organised CV versus a wall of unbroken text.
Specificity
“Experienced coach helping people reach their potential” describes several thousand practitioners simultaneously and is functionally invisible. “I work exclusively with first-time founders in the eighteen months after their first hire” describes one person, and is exactly the kind of detail an answer engine can quote with confidence.
Vague language is invisible to a machine for the same reason it’s forgettable to a human: it describes everyone.
Generative answer engines favour sources that state a clear, narrow claim over sources that gesture broadly at credibility.
— paraphrased from current commentary on generative-engine optimisation
What actually gets cited
In practice, the pages that get pulled into these answers tend to share a shape: a clear statement of who the practitioner works with, a specific method or framework named in the practitioner’s own words, and enough surrounding prose that the claim reads as considered rather than templated. None of this requires gaming anything — it requires writing the way you’d explain your work to a genuinely curious colleague.
A test worth running. Ask an AI assistant your own question — the one a prospective client might ask — and see who it names. If it isn’t you, read what it did cite. The gap between their page and yours is usually the whole answer.
The uncomfortable part
None of this can be bolted on with a plugin or a paragraph of keywords. It requires the site to actually say something true and particular about how you work — which is, not coincidentally, also what makes a human reader trust you. The machines didn’t invent a new standard. They just made the old one, the one great writing always met, impossible to skip.
The answer engine isn’t choosing a keyword. It’s choosing a voice it can quote with confidence.