Asking like the owner.
"Tell me about [business]'s excellent services" invites agreement. Customers ask "is [business] any good?" Test with their phrasing or you will grade yourself on questions nobody asks.
PLAYBOOK · 07 · THE AI ANSWER CHECK · BY MARC KLEINMANN
AI assistants are answering questions about your business. Here is how to hear what they say.
The 15-minute monthly check that shows you what ChatGPT, Gemini, Claude, and Perplexity tell customers about your business. Ten questions, four checks, a printable scorecard, and a skill that runs the whole thing. When I ran it on three real businesses, two came back with the wrong owner.
01 ·Context
One answer instead of ten links, and no error message when it is wrong.
In 2020 a customer typed "estate lawyer near me" into Google and got ten blue links. A growing share of them now ask ChatGPT or Gemini and get a short paragraph naming one or two businesses. Either the machine names you and gets you right, or you were never in the running. This shift is what the acronyms AEO and GEO refer to: answer engine optimization and generative engine optimization, two names for the work of being the business the AI names, correctly, when somebody asks.
The part that can hurt you: Google's list could bury you, but it could not lie about you. A generated answer can. When these systems are missing a fact about your business, they do not say "I don't know." They fill the gap with something plausible and deliver it with full confidence. In my testing, the fact they get wrong most often is the first one a customer would check: who owns the business today.
The check on this page works like a mystery shopper. Same script, every visit, written down, so a change between visits means something. You run it once this week and then once a month, and you know exactly what the machines are telling people about you.
02 ·Architecture
The check has four parts.
Part 01 · Five questions, two phrasingsWritten like a customer, not like the owner.
Ownership, what you sell, reputation, the best-of list, and recency. Each asked twice: once carefully, once the way a real customer types. Casual phrasing finds errors that careful phrasing hides.
Part 02 · Two assistants, freshNew conversations, nothing pasted in.
ChatGPT and Gemini, or Claude and Perplexity. A fresh conversation is the point: you want what the machine tells a stranger, not what it tells someone who just explained the business to it.
Part 03 · Four checks per answerYes or no, no essays.
Named you at all. Facts right, starting with the owner. Cited your actual website. Put you on the best-of list. Four marks per answer and you are done scoring.
Part 04 · One dated file, rerun monthlyThe file is the asset.
Questions, answers, scores, saved with the date. Next month, the same file again: same wording, same assistants. If the questions drift between runs, a changed answer means nothing. Hold them still and a changed answer is a finding.
03 ·Setup
Print the scorecard, or install the skill.
Nothing to install for the manual run: the two copy boxes in the runbook below are the whole thing. If you want paper, the scorecard is one printable page with the questions, the scoring grid, a fix list, and the monthly log.
the scorecard · printable PDF
ai-answer-scorecard.pdf
One page, print it and write on it:
· the five questions, with room for your own wording
· the four-check scoring grid for two assistants
· the fix list, ranked by damage
· the month-over-month log
the same scorecard · HTML
ai-answer-scorecard.html
The identical sheet as a single HTML file, for anyone who
would rather fill it in on screen or print from the browser.
And if you use Claude, the whole check exists as a skill: it builds your ten questions, walks you through collecting the answers, scores them against your real facts, and writes the dated scorecard with the fix list for you. Download the file, then add it to Claude as a skill (in Claude, Settings, then Capabilities, then Skills; or drop it into a Cowork project). If you have never installed one, Writing your first skill shows the whole move.
the whole check · one skill file
ai-answer-check-SKILL.md
All six steps, in order:
· reads your website and freezes the true facts
· builds the ten questions, customer voice, your wording
· walks the collection round in the other assistants
· keeps its own answer honest (a separate fresh Claude
chat answers, not the one holding your answer key)
· scores the four checks and quotes every wrong sentence
· writes the dated scorecard, fix list ranked by damage,
and the rerun instruction for next month
04 ·Decisions
Two calls before your first run.
Decision 01Manual first, or skill first.
If this is your first run, do it manually with the copy boxes below. Reading the machines' answers with your own eyes, in their own confident wording, is worth more than any summary of them. Once you have felt that, let the skill absorb the tedium of the monthly rerun.
Decision 02Which two assistants.
Pick the two your customers most plausibly use, and keep them fixed month to month. ChatGPT and Gemini is the default pair: one is the most-used assistant, the other is wired into the search box your customers already type into. Add a third later if you want; changing the pair mid-stream resets your trend.
05 ·Runbook
The ten questions, then the scoring prompt.
Step one: fill in your business, category, and town, then ask each question in a brand-new conversation in each assistant. Copy the answers somewhere as you go.
the ten questions · fill in and ask
Replace [business], [category], and [town], then ask each question in a brand-new conversation, one at a time, exactly as written. 1a. Who is the current owner of [business]? 1b. who owns [business]? 2a. What products and services does [business] offer? 2b. what does [business] actually do? 3a. What do customer reviews say about [business]? 3b. is [business] any good? 4a. Which [category] businesses in [town] are most recommended? 4b. who's the best [category] in [town]? 5a. What recent changes or news are there about [business]? 5b. what's new at [business]? The a-version is careful phrasing. The b-version is how a customer types. You need both: casual phrasing finds errors that careful phrasing hides.
Step two: paste the scoring prompt into any AI, add your true facts and the collected answers, and it scores the run and writes your fix list. Use an assistant that did not answer the questions, or a fresh conversation at minimum.
the scoring prompt · four checks and a fix list
TASK: Score these AI answers about my business and write me a fix list. CONTEXT: I asked two AI assistants ten questions about my business. Below are the true facts, then the answers. Score every answer with four checks, yes or no: NAMED (it mentioned my business at all), FACTS (details right, starting with the owner), CITED (it referenced my actual website), LISTED (question 4 only: my business appeared on the best-of list). FORMAT: One score table, rows are the five questions, columns are the assistants. Under it, quote every wrong sentence verbatim. Then a fix list: for each error, the likely source (old press mention, thin homepage text, empty description field, stale directory or profile) and the fix, ranked by damage to the business, worst first. Flattering errors rank high too. TRUE FACTS: [owner, what the business sells, locations, anything else a customer might check] ANSWERS: [paste each assistant's answers here, labeled]
Step three: save the whole thing, questions, answers, scores, and fix list, in one file with today's date. Work the fix list. Next month, run the identical questions again and compare.
06 ·Gotchas
The four that catch everyone.
Rewriting the questions each month.
Change the questions and you changed the input, so the changed answer proves nothing. The wording freezes after run one. That is what makes month three mean something.
Trusting a flattering answer.
One business I scanned was credited with an award a direct competitor won. Errors that flatter survive longest, because nobody complains about them. Check the nice claims as hard as the bad ones.
Running it once and moving on.
A single run is a snapshot. The value is the trend: you fix the pages you control, rerun in a month, and watch the scores move. One run tells you where you stand; the loop tells you whether anything you did worked.
07 ·What's next
Most of what you find is fixable in an afternoon.
When I ran this on three real businesses, nearly every error traced to a page the business itself controls: a years-old press quote naming the previous owner, a homepage giving a machine 41 words of readable text, an empty description field the website platform generated. No budget required, just the fix list worked in order.