Inside an AnswerMonk report: an annotated walkthrough
An AnswerMonk report shows your appearance rate on ChatGPT across three search modes, a competitor leaderboard, per-mode leaders, a citation-source breakdown, a winners playbook, and a claims checker — a measurement of how AI answers your buyers, plus the to-do list that falls out of it.
This page walks through each section: what it shows, the question it answers, and what you do with it. If you would rather inspect the real thing first, public example reports live at /reports, and the method behind every number is documented at /methodology.
Appearance rate
The appearance rate is the share of your category's buyer prompts in which your brand was named.
What it shows: the audit generates the prompts your buyers actually ask — comparisons, "best for" questions, alternatives — runs them on ChatGPT across three search modes, and counts how often you appear. The question it answers: when a buyer asks an AI assistant for a recommendation in my category, am I in the conversation? What you do with it: treat it as your baseline. Good looks like appearing consistently on the prompts closest to purchase intent. Bad looks like a category where ChatGPT answers fluently, names your competitors, and never mentions you. Both are common, and both are invisible until you measure.
Competitor leaderboard
The leaderboard ranks every brand ChatGPT named across your prompts, so you can see exactly who owns the answers you are missing from.
The question it answers: who is winning my category inside AI answers, and is it who I expected? Often it is not — brands that trail you in Google can lead you in AI answers, and vice versa. What you do with it: pick the winner whose position you most want to understand, then read the citation sections to see how they earned it.
Per-mode leaders
Search modes disagree, so the report names the brand each mode favors instead of hiding the split behind a blended score.
What it shows: the favored brand in each search mode — search off, model decides, and search forced — side by side. The question it answers: which mode is my problem, and which is already working? The disagreement is not an edge case — from our 300-probe phrasing study: two engine presets agreed on category winners only 22% of the time. What you do with it: stop treating "AI" as one channel. If search-forced answers favor you and the search-off memory answer does not, those are different jobs with different sources behind them.
Citation-source breakdown
This section lists the pages, publications, and directories ChatGPT actually cited — the supply lines behind every recommendation.
What it shows: the sources cited across your prompts, with the ones carrying the winning brands called out. The question it answers: where does visibility in my category actually come from? The wider pattern matches what per Muck Rack: 84% of AI citations come from earned media — coverage and third-party mentions, not brand homepages. What you do with it: compare the cited-source list against your own footprint. Every cited source you are absent from is a named, addressable gap, not a theory.
Winners playbook
The playbook shows which sources carry the winning brands, and where your brand could get listed.
What it shows: get-listed targets drawn from the winners' actual citation footprint, plus the on-page traits that cited pages share. From our crawl of citation-winning pages: FAQ blocks and question-form headings recur among cited pages (correlational). The question it answers: what, concretely, do I do next? What you do with it: work the list. Some targets are a submission form away; others are earned coverage that takes months. The report tells you which is which — the working of the list is still marketing.
Claims checker
The claims checker flags wrong facts ChatGPT states about your brand, with per-claim provenance.
What it shows: each checked claim, whether it holds up, and the provenance trail behind it. The question it answers: is AI misinforming my buyers about my product, pricing, or availability? What you do with it: fix the upstream source. You cannot file a support ticket with a language model, but you can correct what it reads — and the provenance points you at the likeliest origin.
What a report is — and is not
A report is a measurement plus a to-do list, not a magic ranking button.
Nothing in it promises an outcome, because nobody can honestly promise how an AI engine will answer next month. What it gives you is the state of play — who wins, where, and on the strength of which sources — and a work list ordered by what the winners actually did. If you want ongoing help, the $19/month tier adds plain-language action plans, a living knowledge base tracking Schema.org, Google, OpenAI, and Perplexity guidance, a source-placement agent, and WhatsApp lead capture. Browse the public examples at /reports before you run your own.
Run your free AI visibility audit →
Frequently asked questions
Where can I see a real AnswerMonk report?
Public example reports are published at answermonk.ai/reports. They are full reports on real categories, so you can inspect every section before running your own. Your own audit is free, takes 3-8 minutes, and requires no signup.
Why do ChatGPT's search modes name different winners in the same category?
The same model answers differently depending on whether it searches the web or answers from memory, so modes routinely disagree. The split is measurable — our 300-probe phrasing study: two engine presets agreed on category winners only 22% of the time. That is why AnswerMonk reports per-mode leaders instead of a single blended score.
What does the claims checker in an AnswerMonk report flag?
It flags wrong facts ChatGPT states about a brand, such as outdated pricing or features that do not exist. Each flagged claim comes with per-claim provenance, so you can trace the likely origin and correct the source.
Does an AnswerMonk report tell me what to do, or just measure?
Both. The free report includes a winners playbook showing which sources carry the winning brands and where to get listed. The $19/month tier adds plain-language action plans, a living knowledge base, a source-placement agent, and WhatsApp lead capture.
How long does an AnswerMonk report take to generate?
Results arrive in 3-8 minutes. You enter a URL at answermonk.ai with no signup, and the audit generates your category's buyer prompts, runs them on ChatGPT across three search modes, and returns the full report.
Run a free AI visibility audit for your brand at answermonk.ai