Learn · Field notes
Plain notes for business owners on how AI assistants choose who to recommend, what they read about you, and what you can actually change. No promises nobody can keep.
21 notes · newest first
Most share-of-answers charts measure something else and label it local. The number that matters is not the market’s — it is yours.
· updated · 8 min read
Start here if anyone has shown you a chart of “AI market share”.
Research7 min
Five buyer questions, three assistants, one run each. Nearly half the practices we checked did not appear in a single answer, and the sources behind the answers were mostly not the practices’ own sites.
Playbooks6 min
An assistant can only repeat what is written down somewhere it reads. For a practice that means a short list of plain facts, stated on your own site and agreed with everywhere else.
AI search basics5 min
A patient choosing who to trust with their face asks an assistant first. In the few seconds before it answers, it reads review pages, directories, list sites and, if they say anything, practices’ own pages.
AI search basics8 min
The honest answer has four steps, none of them exotic, and one uncomfortable caveat: nobody can promise you a place in an answer, including us.
AI search basics8 min
A fair question, usually asked by somebody who has been sold a rebrand before. The overlap is real. So are five differences that change what you do on Monday.
Playbooks8 min
Every objection to publishing prices is about competitors. Every consequence of not publishing them falls on the customer who wanted to choose you.
Playbooks8 min
Your customers know you cover the next town over. Nothing an engine reads knows it, because a map graphic and a radius are not sentences.
AI search basics8 min
Between somebody asking and your name appearing there are a few seconds of reading. Knowing what gets read explains almost every odd answer you have seen.
AI search basics8 min
The most frustrating sentence in this whole subject. You control your website completely, and it is the source with the least say in how you are described.
Playbooks8 min
Structured data is the one place you can state your facts in a form built to be copied. Here is the whole block, with the reasoning for every line in it.
Playbooks9 min
Nobody sets out to hide from the engines. They inherit a template, or ship a staging file, and quietly stop being readable by the thing writing the answers.
Playbooks8 min
Directories will always be cited. The question is whether anything of yours is cited alongside them — because the sources decide the description, and you only control one.
Research8 min
Every tool publishes what it measures. Almost none publishes what it misses — which is the half that decides whether the first half means anything.
Research8 min
Your friend asked ChatGPT and you were there. You asked and you were not. Neither of you is wrong — and the disagreement is the most useful thing on the page.
Research8 min
A visibility score is only as good as the questions behind it. Here are ours in full — including the rules that cost us points we could have quietly claimed.
Research9 min
A number you cannot take apart is not a measurement, it is a mood. This one takes apart into seven pillars — ninety points of observation and ten of forecast.
AI search basics8 min
Search gave you ten chances to be chosen. An answer gives you one paragraph naming two or three businesses. There is no page two to grind up from.
AI search basics8 min
One file at your domain root, stating what you do, where you do it, and who you serve. It takes about an hour.
Playbooks8 min
The sources an engine leans on are neither secret nor exotic. That is what makes this a task you can finish rather than a strategy you maintain.
Playbooks8 min
If four sources say you close at five and only your website says nine, the answer says five. Here is how to work out which source to fix first.
The guide behind fix 2 in the sample report: the wrong hours.
Your own answers
These notes say what usually goes wrong. A scan shows what AI says about you: we ask ChatGPT, Gemini and Perplexity your customers’ questions and show who they name.
Not ready? See a sample report first.