AI Search Optimization
Free AI Visibility Audit:See Whether ChatGPT, Gemini and AI Overviews Mention You
People now ask ChatGPT, Gemini, Perplexity and Google AI Overviews who to hire and who to trust. An analyst asks those engines the questions your buyers ask, then sends you a dated log of who is mentioned, who is cited, which sources were shown, and what we would change.
Analyst-Run, Not Automated
Strategy by Anthony Will, Founder & CEO · 13+ years in ORM · 5,000+ clientsRated 5.0 · read 228 client testimonials(opens in a new tab)- What you get. A dated log of how six AI engines answer your buyers' questions, whether you are mentioned or cited, which sources they showed instead, which competitors appear, and a fix list. What the audit covers →
- A human audit, not a score. Answers vary from run to run, so we run each prompt three times and report the runs. A single number would hide exactly the variation you need to see. Why no score →
- Check yourself today. You can spot-check your brand in ChatGPT for free in a few minutes. The steps are below, along with the limits of a spot check. Do it yourself →
What is an AI visibility audit?
The short answer
An AI visibility audit tests whether AI assistants mention, cite or recommend your brand when people ask the questions your buyers ask. Ours is run by an analyst, not a script: a fixed set of prompts across ChatGPT, Gemini, Claude, Perplexity, Copilot and Google AI Overviews, logged with dates, with every cited source recorded. You get the log, the gaps, and what we would change. It is free and carries no obligation.
What the audit covers, and what it does not
What the audit covers. An analyst asks six engines the questions your buyers ask: ChatGPT, Gemini, Claude, Perplexity, Copilot and Google AI Overviews. The prompts fall into four classes: identity (who is this company), legitimacy (can it be trusted), comparison (how does it stack up against named alternatives) and recommendation (who should I hire or buy from). Where a product lets us switch web search on and off, we run both, from logged-out US sessions, and every prompt is run three times. The limit: this is a sample of answers on the days we ran them, not a census of everything these engines could say to every person, and logged-out US sessions do not show what a personalized or non-US session would return. The audit is the first step of our AI search optimization services and of our AI reputation management work, depending on what it finds.
What the report contains. For each engine and prompt class you get the mention rate (how many of the runs named you), the citation rate (how many showed a link to your site), the list of sources each engine cited, how you were characterized in your own words and in theirs, the competitors that appeared, and a fix list ordered by what we think matters most. The limit: the report shows what the engines said and which sources they showed, but engines do not disclose why they chose a source, so any explanation is our analyst's reading of the evidence, and we label it that way.
Why we do not give you a score. Answers vary from run to run, and a single number hides that variance, so a score would look precise while measuring very little. We report the runs instead: for each prompt, what each of the three runs said and whether it named you. The limit: you do not get a tidy figure to paste into a slide, and we do not compare you against an industry average, because we have no verified one to offer.
How to check your brand in ChatGPT yourself, free. First, open ChatGPT in a private browser window, signed out if the product lets you, so your own history does not color the answer. Second, ask an identity question such as "What is [your brand]?" and then a legitimacy question such as "Is [your brand] legitimate?". Third, ask a comparison question such as "How does [your brand] compare to [a competitor]?" and a recommendation question such as "Who are the best [your category] in [your city]?". Fourth, repeat each question a few times, with search on and off if the product offers both, and write down every answer, every source it shows, and the date. The limit: one person on one afternoon sees a handful of answers, which is a spot check and not an audit. Our guide to how to monitor your brand in AI search explains the full method.
What happens after the audit. You get the report and a call to walk through it. You can use the fix list yourself, hand it to your own team, or ask us to scope the work, and anything we scope is quoted separately in writing before you decide. If you want ongoing monitoring after the audit, that is our AI reputation monitoring service. The limit: the audit is a point-in-time picture, and it does not by itself change what any engine says about you.
Reputation audit or AI visibility audit? Our free reputation audit looks at what Google results and reviews say about you, while this audit looks at what AI answers say about you and whether they cite you at all. If your worry is a damaging search result or review, start with the Google side; if your worry is whether ChatGPT or Gemini names you when a buyer asks, start here. Many businesses want both, and the two are separate requests. The limit: neither one predicts the other, because an AI answer can differ from what Google shows for the same question.
Why It Matters
You cannot see AI answers in a rank report
An answer is not a list of ten links
When a buyer asks an assistant who to hire, the engine composes an answer and names a few companies, or none. You are either in that answer or absent from it, and a search rank report does not show which.
Each engine can say something different
The six engines draw on different sources and behave differently, so you can be named in one and missing from another. The audit logs each engine separately so you can see where you stand in each.
Answers change, so one check proves little
The same question can produce different answers on different runs. That is why we run every prompt three times and report all three, instead of reporting one lucky or unlucky answer.
You see who is named instead of you
When an engine skips you, it usually names someone else. The report lists the competitor set and the sources behind it, so you know who and what the engines are leaning on in your category.
Engines We Check
Six engines, each logged for what it actually shows
ChatGPT
ChatGPT can be used with web search on or off, so we test both where the product allows it. We log every source it shows with each answer. We cannot see why it chose them, and we do not claim to.
Gemini
Google's help documentation says Gemini Apps may show a Sources button with links, and that not every response has sources. So we log whether sources appear for each prompt, which ones, and note when there are none.
Claude
Anthropic's help documentation describes web search as a Claude feature for factual lookups. We test with search on and off where it is available and log whatever sources Claude shows, including when it shows none.
Perplexity
Perplexity's help documentation says each answer includes numbered citations that link to the original sources. We record every numbered source for each run, so you can see which pages it leaned on and whether yours was one of them.
Copilot
Microsoft's documentation says that with web search enabled, Copilot may fetch information from the Bing search service, and that Copilot Chat shows the search queries it ran in its citation section. We log the queries and sources shown for each run.
Google AI Overviews
Google's documentation says AI Overviews and AI Mode link to supporting pages, that a page must be indexed and eligible for a snippet to be one, and that no additional technical requirements apply. AI Overviews also appear only for some queries, so we log when one does not appear.
How It Works
How the free AI visibility audit works
- 01
You request the audit
Free to start.Tell us your brand, your category and who your buyers are. We tell you the delivery date when you request it.
- 02
We build the prompt list
Your questions, not ours.An analyst writes the identity, legitimacy, comparison and recommendation prompts your buyers would plausibly ask, and you can add or remove prompts before we run them.
- 03
We run every prompt, three times
Logged, not summarized from memory.Each prompt goes to each of the six engines, with search on and off where the product allows it, from logged-out US sessions. Every answer and every source shown is saved with its date.
- 04
The analyst reads the evidence
Evidence first.We tally mention and citation rates, read how you were characterized, list the competitors and sources that appeared, and write the fix list. Where we are interpreting rather than reporting, we say so.
- 05
You get the report and a call
No obligation.You receive the log and the report, and we walk through them on a call. Any work beyond the audit is quoted separately in writing, and only if you choose it.
Free to find out. No obligation.
Tell us your brand and we will tell you the delivery date. Any work after the audit is quoted separately in writing, and only if you choose it.
Honest Timelines
What happens, and when
We do not publish a fixed turnaround here. We tell you the delivery date when you request the audit, so you get a real date rather than a generic range.
We tell you the delivery date when you request it, based on the number of prompts your category needs.
You see the prompts we plan to ask and can add or remove any of them.
Every prompt is run three times on each engine, and the analyst reads the logged answers. Nothing is summarized from memory.
You receive the log, the gaps and the fix list, and we go through them with you on a call.
Further work is scoped and quoted separately in writing. No one can guarantee what a specific engine says about you after any piece of work.
Why We're Different
Analyst-run audit vs an AI visibility checker
Who runs it
AI Visibility Checkers
Software runs the prompts on a schedule
Reputation Resolutions
An analyst runs and reads the prompts
What you receive
AI Visibility Checkers
Output format varies by tool
Reputation Resolutions
A dated log, the gaps, a competitor set and a fix list
Variation between runs
AI Visibility Checkers
Handling of run-to-run variation varies by tool
Reputation Resolutions
We run each prompt three times and report all three
Interpretation
AI Visibility Checkers
You decode the output yourself
Reputation Resolutions
An analyst explains what was found, and labels interpretation as interpretation
Repeat checks
AI Visibility Checkers
Can re-run automatically
Reputation Resolutions
A point-in-time audit; ongoing monitoring is a separate service
Cost to start
AI Visibility Checkers
Varies by tool and plan
Reputation Resolutions
Free, with no obligation
Who runs your case
Senior specialists, no junior handoffs
Reputation Resolutions is run and managed by a world-class team of online reputation management experts. Your case is handled by senior, multidisciplinary specialists: removal strategists who know each platform's rulebook, SEO and content experts who rebuild your search results, legal partners for the matters that need them, veteran PR professionals, and AI-search specialists who help you control what LLMs like ChatGPT say about you. There are no junior handoffs and no learning on your case, and every person here treats your name as if it were their own.
Get Started
Get Your Free AI Visibility Audit
Tell us your brand and who your buyers are. An analyst will ask six AI engines the questions they ask and send you a dated log of who is mentioned, who is cited, and what we would change, free and with no obligation.
Free & Confidential Consultation
Tell us what’s going on (required)
Prefer to talk? Call 855-239-5322
AI Visibility Audit FAQs
The Free AI Visibility Audit, Answered Honestly.
Straight answers on cost, which engines we check, why runs vary, competitors, scores and what comes next.
Yes. The audit is free and carries no obligation: you receive the report and a call to walk through it whether or not you ever work with us. If you later want help acting on the fix list, that work is scoped and quoted separately in writing before you decide, and you are free to take the fix list to your own team instead.
They look at different places. Our free reputation audit looks at what Google results and reviews say about you. The AI visibility audit looks at what AI assistants say about you when a buyer asks them a question, and whether they cite your site at all. If your concern is a damaging Google result or review, start with the reputation audit. If your concern is whether ChatGPT, Gemini or Google AI Overviews name you, start here. Neither predicts the other, because an AI answer can differ from the search results for the same question.
Six: ChatGPT, Gemini, Claude, Perplexity, Copilot and Google AI Overviews. Where a product lets us switch web search on and off, we test both, from logged-out US sessions. The limit is that engines change their products often, and a logged-out US session does not show what a signed-in or non-US session would return, so the report states the dates and conditions of every run.
We do not pretend they do not change. Every prompt is run three times on each engine, and the report shows all three runs, so you can see whether you were named in all of them, some of them or none. A single run can mislead in either direction. Even three runs is a sample, not proof of what an engine will say tomorrow.
Yes. The report lists the competitors that appeared in the answers, how often, and which sources the engines showed alongside them. The limit is that this reflects the prompts we ran. A competitor that appears for a question we did not ask will not appear in the report, which is why you can review and add to the prompt list before we run it.
Yes, and it is worth doing today. Open ChatGPT in a private browser window, signed out if the product lets you. Ask who your company is, whether it is legitimate, how it compares to a named competitor, and who the best providers are in your category and city. Repeat each question a few times, with search on and off if available, and write down every answer, every source shown and the date. That is a useful spot check, but it is a handful of answers from one person on one afternoon, not an audit. Our guide to how to monitor your brand in AI search explains the full method.
No, deliberately. Answers vary from run to run, and a single number hides that variance while looking more precise than the data is. We report the runs instead: for each prompt and engine, what was said and whether you were named. If you need a number for an internal slide, you can calculate your own mention rate from the log, and we will show you how.
You get the report and a call. From there the choice is yours: use the fix list yourself, hand it to your own team, or ask us to scope the work. Anything we scope is quoted separately in writing, and no one can guarantee that a specific engine will mention or cite you afterward. If you want ongoing monitoring after the audit, that is our AI reputation monitoring service, and it is a separate decision from the audit.
The brand, category and questions you give us are handled confidentially, and we use them to run your audit. If a question involves something sensitive, tell us when you request the audit and we will talk through how to handle it.
Still not sure if your situation qualifies?
Get a straight answer from a senior specialist in one call: free, confidential, and you'll know exactly where you stand before you decide anything.
Keep exploring
Related services and guides
Related services
