Write for the machine that quotes you

An AI doesn't read your page the way a person does. It doesn't savour your opening line or appreciate the long build to your point. It scans for a clean, self-contained statement that answers a question — and lifts it. If your best idea is buried three clauses deep in a beautiful sentence, the machine walks past it.

Writing to be cited is a different discipline from writing to be admired. Both can be good. Only one gets quoted.

Make it extractable

Extractable means a passage stands on its own. Pulled out of the page and dropped into an answer, it still makes sense and still credits you. That favours a few specific habits:

1. Answer the question in the first sentence of a section, then elaborate. Don't make the machine wait for your conclusion.
2. One idea per paragraph. A paragraph that argues three things is hard to quote cleanly.
3. Name things plainly. Use the words people actually search and ask, not your internal cleverness for them.

The machine quotes the sentence that survives being taken out of context. Write more of those.

Depth is what earns the citation

Extractability gets you read; semantic depth gets you trusted. LLMs cite sources that genuinely cover a topic — context, specifics, the surrounding why — not thin pages stuffed with keywords. Structure that depth with real headings, FAQ-style questions, and proper schema so the meaning is legible to a crawler as well as a reader.

For a property studio, that might mean a neighbourhood page that actually answers "what's it like to live here?" in clear, quotable pieces — schools, the walk to the station, the morning light — rather than a brochure of adjectives.

You're no longer only writing for the person who lands on the page. You're writing for the system that decides whether to mention you at all. Clarity was always good manners. Now it's also distribution.