AI Visibility Audit:
Who Does the AI Name Instead of You?

Last updated: October 14, 2026

Quick answer

An audit of your AI visibility records which company ChatGPT, Google AI Overviews, and other assistants recommend for your buyers' questions, then shows you the facts those assistants can find about the competitor they named instead of you. We test crawler access, fact consistency across your listings, how quotable your pages are, and citations on at least six AI platforms, then compare your results with your three closest competitors.

AI visibility audit: a magnifying glass over a brand citation report

Key takeaways

  • The audit measures whether Gemini, Perplexity, Google AI Overviews, Google AI Mode, and other assistants name you, not just whether you rank in the organic results.
  • AI tools tend to recommend businesses whose facts agree in several places, not simply the company with the most backlinks.
  • Free tools give you a one-time AI visibility score; your real score comes from dated answers tracked over weeks, and only an audit shows why the gap exists.
  • Search rankings and AI answers share some signals (schema markup, authority, clear pages), but AI search engines weigh verifiable facts and what other sites say about your brand more heavily.
  • Service firms whose buyers research before they call (law firms, specialty contractors, consultancies, clinics) gain the most when assistants start naming them.
  • The right order is crawler access, then fact accuracy, then pages an assistant can quote, then internal linking.
  • In our published records, AI citation counts rose threefold to sixfold between July 1 and October 13, 2026, and inquiries follow later, on the buyer’s schedule.

What is an AI visibility audit, and how does it differ from an AI SEO audit?

An AI visibility audit records the AI answers Perplexity, Gemini, and other assistants give when buyers ask who to hire, then investigates why an assistant finds one company easier to describe and recommend than another. It is not a ranking report, and it is not a one-time number from free tools.

Illustration of a business owner reading an AI answer that names four other marketing companies first and lists his own brand eighth

We run the same buyer questions across ChatGPT, Gemini, Perplexity, and other AI platforms and record the date, the full reply, and the company named, so the sentiment of what each assistant says about you is on file too. Then we trace the supporting evidence, including authorship, matching company details, and the pages an assistant can read and quote. According to Search Engine Land's AI search optimization framework, identifying the exact authority gap between a brand and its AI-cited competitor is the step most generic marketing audits skip entirely.

A proper audit also benchmarks you against three real competitors the assistants are already naming, scoring each one against yours on the same scale in a side-by-side competitor comparison. That comparison is what turns a vague "we're not showing up in AI" complaint into a fixable list.

Why does this matter for my company?

This matters if your buyers research services before calling you, because AI tools are increasingly the first stop in that research, and traditional SEO rankings alone don't guarantee you get named there. Without one, you're guessing at why a competitor gets named instead of you.

Consider what changes when a company closes that gap:

  • An immigration law firm serving Seattle and Vancouver went from 4 ChatGPT visits a month in July 2026 to 75 by September, after a content rebuild that added named authorship and sourced its claims against official government references.
  • A special education law firm in California saw website inquiries rise from 41 to 74 over the same four-week window year over year, tracked in its own analytics.
  • A wellness spa in Montana went from 43 AI citations on July 1, 2026, to 158 by October 20, measured across every major AI platform in Ahrefs.

None of these businesses changed what they sell. As the data shows, they changed what the AI could find about them.

How does ChatGPT affect my Google rankings and sales?

ChatGPT does not directly change your rankings. It can, however, shape the shortlist before a buyer visits your site, which affects the inquiries you receive even when your rankings stay the same. A buyer who gets a confident, cited reply from an assistant may never run a search at all.

This matters more every quarter. Search Engine Land's GEO measurement analysis found that the large majority of businesses still have no visibility into how often assistants mention them, even as buyer behavior shifts toward asking an assistant first. If an assistant names your competitor, that inquiry never reaches your contact form, and your organic ranking never gets the chance to compete for it.

Our cell tower lease consultancy case makes that connection concrete. Monthly search clicks rose from 71 in July 2026 to 111 by September, and the firm held top-three map placement in 43 of 47 tracked local searches that August. Organic rankings and AI mentions improved together because the underlying fixes (clearer pages, verifiable facts, consistent company details) serve both.

What's the difference between search rankings and AI recommendations?

Ranking work aims for position in a list of links; answer engine optimization aims to make your business the fact an assistant chooses to state and cite in a direct answer. The mechanics overlap, but the scoring doesn't.

Factor Search rankings AI answers
Primary goal Rank on page one Get named in the reply
Key signal Backlinks, query match Verifiable facts, named authorship, brand mentions
Measured by Position tracking Citation counts across AI platforms, dated
Content format Optimized for search intent Optimized to be quoted directly
Local signal Map pack position Listing consistency across the web

One piece of original research covering 75,000 brands found that brand mentions across the web tracked AI Overview visibility more closely than backlinks did, a signal traditional ranking audits rarely measure. For a deeper look at how these two disciplines diverge in practice, see this comparison of AI SEO versus traditional SEO.

If your main question is "why do AI assistants recommend someone else," start here. If the question is "why don't we rank higher in organic results," start with a traditional SEO audit.

What makes an AI assistant recommend a firm?

Assistants are more likely to recommend a company when its facts agree in several places, on its own website and across the wider web. The authority signals they look for are a named person behind the writing, matching details across directories and reviews, and references to the brand on third-party sources.

The immigration law firm case shows this in detail. Across its whole website, it pointed to official government sources behind its pages just 3 times before its August 2026 rebuild. In one week, it added named authorship and the schema markup that tells search engines who wrote each page. By the end of that week, source references across the website had grown to 69. Google AI Overview citations for the firm rose from 21 on July 1, 2026, to 90 by October 11.

The same pattern shows up outside legal services:

  • A paint protection film specialist in England went from 35 AI citations on July 1, 2026, to 109 by October 13, with Google AI Overview citations climbing from 11 to 67 in the same window.
  • A bathtub reglazing company in Southern California grew website form inquiries from 12 to 21 across the same July-to-September period year over year.

A crawler that can't reach your pages can't read any of this. We test each major AI crawler by name against your key URLs, because an assistant can't quote a page its crawler never saw. Our own page passed that test for all 8 tracked AI crawlers plus a standard browser on October 11, 2026.

Which companies benefit most from this audit?

Service firms whose buyers research before they call benefit most: law firms, medical and dental practices, specialty contractors, consultancies, and e-commerce stores selling considered purchases. If your buyer compares options before reaching out, an AI assistant is likely already part of that comparison.

Businesses with less to gain include:

  • Businesses competing purely on price, where buyers don't ask an assistant for a recommendation.
  • Businesses that only serve walk-in or impulse demand with no research phase.
  • Businesses already guaranteed the work through contracts or referrals, with no open buyer search to influence.

For a local business with multiple locations, consistency across every listing matters even more. See how multi-location businesses manage AI search consistency across locations, since one mismatched address or phone number can quietly undercut otherwise strong directory profiles with AI tools.

What should the audit include?

A complete AI visibility audit includes at least nine checks against your three closest competitors. We examine crawler access and technical health, company facts, how quotable your pages are, and actual citation counts across multiple AI platforms. A tenth check covers your Google Business Profile and local map visibility if you serve a defined area.

The nine core checks should cover:

  1. Crawler access. Whether each major AI crawler, tested individually, can actually reach your most important pages.
  2. Page cannibalization. Where two of your own pages compete for the same search, and which one should own it. See how keyword cannibalization gets diagnosed in more detail.
  3. Fact consistency. Whether your company details and credentials are present and match across your website, Google Business Profile, directories, and review platforms.
  4. Quote readiness. How clearly your pages respond to buyer questions and how well they survive AI extraction, judged against a fixed set of tests. This overlaps with answer engine optimization services.
  5. Citation sources. The cited sources, yours or a competitor's, the assistants actually use for your buyers' questions.
  6. Internal architecture. Gaps in internal linking and page structure that keep an assistant from reaching the pages that bring you work.
  7. Conversion scoring. Pages that get visits but don't turn them into inquiries, scored against a fixed benchmark. This is the same lens used in a conversion rate optimization audit.
  8. Named mentions. How often AI tools name your brand, your competitors, or neither, across a fixed prompt set of branded prompts and category questions.
  9. Competitive benchmarking. An AI visibility score out of 50 on verifiable facts, for you and each of your three competitors, showing the exact gap against their average scores.

Search Engine Land's brand visibility measurement guidance notes that the audits getting results treat AI answers as a measurement discipline with dated evidence, not a single pass/fail grade. That matches what our own case records show: every count here is tied to a specific date and source, not a one-time snapshot.

Phil Belleville, the person you'll talk to

Phil Belleville, our founder, has worked in search marketing every working day since 1995. Your audit is the first step of his Rank Intelligence Engine, the system he built to diagnose why a WordPress website misses both the top organic results and AI answers.

Phil reads the evidence behind every finding personally and decides what it means before it goes in your report. He led the diagnostic work behind every result cited here, from the immigration firm's AI Overview citations to the AI citation counts for the Montana spa and the paint protection film specialist in England. You talk to him directly on the strategy call, and he leads your audit from there. A USA-based team of nine installs any approved changes, with every page reviewed by Phil before it goes live.

Book your audit call with Phil

No account access needed. No obligation to buy.

How is the price set?

The price of an AI visibility audit depends on the size of your website and how many buyer questions and AI platforms get tested, and you get that number in writing after a free strategy call, before paying anything. There's no published flat rate because a five-page local website and a 400-page e-commerce catalog require very different testing scope.

What the quote covers, specifically:

  • The key pages that bring you work, named individually.
  • The three competitors you're benchmarked against.
  • The delivery schedule for the final report.
  • The search engines and assistants included in testing.

If a provider or marketing agency quotes you a flat number before seeing your website or naming your competitors, that's worth questioning. For context on how pricing conversations should work, see this breakdown of AI SEO pricing and proposals.

Can I test this myself, or do I need help?

You can type a few prompts about your industry into an assistant yourself, or try free tracking tools, but that single-day test won't tell you what's actually driving the result, and a real audit needs dated, repeated testing across platforms to produce a usable finding. A do-it-yourself test is a starting point, not a substitute.

Here's the gap in practice. We recently ran a prompt set of 26 fixed questions on six major AI platforms over the seven days to September 16, 2026. The answers named 18 agencies, including 5 that were not on our original competitor list. A single manual query would have missed most of that picture entirely, along with the sentiment of each reply, and the same question asked again the next day from a different account can return a different response.

Do it yourself if you just want a rough sense of whether your name comes up at all. Get help if you need to know why a specific competitor gets named, what facts about your company the assistant can't find or reconcile, and what to fix first. Search Engine Journal's guidance on auditing a brand in AI search recommends exactly this layered approach: quick manual checks to spot a problem, structured testing to diagnose it.

What happens if AI assistants never name my company?

If buyers never see your own name in those replies, they may contact a competitor before your website has any chance to make its case. This is a silent loss: no referral traffic, no bounce in your reports, just inquiries that never arrive.

Before the gap closed, the immigration law firm received only 4 ChatGPT referral visits a month through July 2026. By September, after the fact and authorship fixes, that figure had reached 75 a month. Every month that gap stays open is a month of inquiries routed somewhere else.

How do I get my website to appear in AI search results?

Three things have to be true before your website can appear in AI search features and assistant replies. The crawler must be able to read your pages, your company facts must agree across the web, and your pages must give replies an assistant can quote directly. All three need to be true at once; missing one undermines the other two.

The order matters:

  1. Fix crawler access first. A page an AI crawler can't reach can't be cited, no matter how good the content is. Facts that only load through JavaScript, rather than sitting in the raw HTML, and pages missing from the Bing index are just as invisible to the assistants that rely on them.
  2. Fix facts second. Mismatched addresses, unverifiable credentials, or inconsistent company details give an assistant a reason to distrust everything else on the page. One audit of a South Florida mobile outboard repair company found an industry award and a certification, appearing 582 times across the site, that no public source could back up; removing both came first. For more on why this matters, see E-E-A-T's role in AI search.
  3. Fix content third. Rewrite key pages so the first sentence of each section answers a buyer question directly, with named authorship, original data, and claims sourced to primary sources. This is the core of generative engine optimization services.
  4. Fix internal links last. Once the facts and pages are right, internal linking helps an assistant's crawler find the pages that matter.

Google's AI search guidance follows a similar sequence: fix access before content, because rewrites on unreachable pages produce nothing measurable.

How long before you see results?

Timing depends on your starting point, but the dated records on our case studies page give a fair yardstick: AI citation counts that rose threefold to sixfold between July 1 and October 13, 2026. Citation tracking tends to move faster than inquiry volume because assistants re-crawl and re-evaluate sources on a shorter cycle than buyers take to act on a recommendation.

Checkpoints matter more than a single end date. A sound program measures:

  • 6 weeks: early signal on whether AI citation counts are moving at all.
  • 12 weeks: a clearer read on ranking movement and AI visibility trend, not just a single data point.
  • 6 months: inquiry volume and revenue impact, measured in your own reports.

The paint protection film specialist's climb from 21 to 89 AI citations over roughly 14 weeks (July 1 to October 11, 2026) sits inside that normal window, not an outlier result.

How do I know if my competitors are winning in AI search?

A competitor is winning when the same dated set of buyer prompts names it more often than you across several AI search engines. We also test whether the facts an assistant can find about that competitor are stronger than yours. A single manual query, or a one-off report from free tools, isn't evidence; a repeated, dated test is.

A proper competitor comparison includes:

  • The same prompts asked across six or more AI platforms, not just one assistant.
  • A score out of 50 for you and each of your top three competitors on verifiable facts.
  • A log of the cited sources behind each answer that named a competitor.

Research on optimizing visibility in generative engines points to the same conclusion about generative engine optimization: businesses that win citations consistently are the ones whose facts can be verified across multiple independent sources, not the ones publishing the most pages. That's the gap this kind of benchmark is built to expose.

How to start, and what stays yours

Start with the AI visibility audit as a standalone paid project, then decide on ongoing work once you've seen the findings; there's no obligation to continue. Pricing and timing depend on your website's size and the number of questions and platforms tested, quoted in writing after a free call, before you pay anything.

This fits WordPress-based service businesses and online stores whose buyers research before they call. It's not built for businesses shopping purely on price or looking for a guaranteed ranking position, since no legitimate provider can guarantee that an AI will mention you.

What you get and what you keep:

  • A written report naming the page and the remedy for every finding, in the order the work should happen, plus workbooks your team or developer can follow.
  • The full audit report stays yours whether or not you hire the firm for implementation.
  • If you do move forward, you keep your logins and account access, and every changed page gets a before-and-after record so anything can be rolled back.
  • Work runs month to month with no long-term contract and 30 days' notice to stop.
  • Every change in scope gets installed and verified on your live website, including schema markup, FAQ schema, and meta tags, though no one can guarantee a specific AI mention in return.

Book your audit call with Phil

30 minutes with Phil. Written scope and pricing before paid work begins.

Frequently asked questions

Can't I just ask an AI assistant myself?
You can, but the reply you get on a given day may be inaccurate, incomplete, or negative in sentiment, even describing services a company never offered. A real audit runs a fixed question set across six AI platforms on recorded dates and counts how often each competitor gets named, which is a different, more reliable signal than one manual query.

Why do I get a different reply every time?
Assistants personalize replies per user and can change what they say from one day to the next for the same question. That's why a credible audit keeps every reply dated and only calls something a finding once it shows up repeatedly.

What if I already have a marketing agency?
Keep your current marketing agency for rankings and traffic diagnostics, and hand them the AI report to work from. This audit is an independent second opinion on whether AI platforms recommend your brand, separate from traditional ranking and traffic audits.

Does the audit include the fixes?
No. The audit hands you the page and the recommended fix for every finding. Making the changes is a separately agreed next step, and Phil leads that implementation work if you choose to move forward.

Is AI replacing traditional search?
Not entirely, but AI tools are increasingly the first stop for buyers comparing options before they run a regular search. Brands that show up in both places capture buyers at every stage of that research.

What's the difference between ranking work and AI optimization in practice?
Ranking work improves position through relevance, page quality, links, and technical accessibility. AI optimization, also called GEO or AEO, makes your brand easier for an assistant to verify and cite through clear authorship, consistent details, and sourced claims.

Do you guarantee my company will be mentioned?
No. Every change within the agreed scope gets installed and verified on your live website, including structured data, but no legitimate provider can guarantee that any assistant will choose to mention a specific company.

What happens after the audit is delivered?
Phil walks you through the roadmap on a review call, explaining what each finding means and the order fixes should happen in. From there you decide whether to implement the changes yourself, hand them to your existing developer, or have the audit team install and verify them.

References

  1. Search Engine Land, How to Optimize Content for AI Search Engines: A Step-by-Step Guide
  2. Search Engine Land, Measuring GEO: What's Trackable Now and What's Still Missing
  3. Search Engine Land, How to Measure Your AI Search Brand Visibility and Prove Business Impact
  4. Search Engine Journal, Brand Protection in AI Search: How to Audit and Defend Your Brand's Identity
  5. Google Search Central, Top Ways to Ensure Your Content Performs Well in Google's AI Experiences
  6. Princeton University, GEO: Generative Engine Optimization (arXiv)