Flagship service • Manual AI visibility audit

Manual AI Visibility Audit: Know Where Your Brand Appears

Understand how AI search answers mention your business, which competitors are recommended and what sources are cited. Our manual investigation turns buyer-question research into verified findings and a practical improvement plan.

Scopes and platform coverage are agreed before work starts. No placement or performance guarantees.

Direct answer

What is a manual AI visibility audit?

A manual AI visibility audit is a structured review of how a brand appears in selected AI-assisted search answers. The researcher tests relevant buyer questions, documents mentions and recommendations, checks competing businesses and citations, and investigates the website issues that can actually be verified.

It is different from typing your company name into a chatbot once. An audit needs a defined market, a reproducible question panel and a clear separation between what the AI response showed and what might explain it.

If you want to check your own brand before buying a service, follow our step-by-step AI visibility checking guide. For the meaning of mentions, citations and recommendations, start with what AI visibility means.

What we investigate

The five parts of an AI search visibility audit

These checks are connected but not interchangeable. Your scope determines the markets, brands, prompts and sources that matter.

01 / Search intent

Buyer questions and prompts

We organize questions by discovery, comparison, shortlist, fit and branded intent. A realistic set may include “Which firms handle multi-state tax filings?” or “What are alternatives to this provider?” The questions come from business context and relevant keyword research—not a random list designed to make your brand appear.

02 / Answer observations

Brand mentions versus recommendations

A mention simply names a company. A recommendation presents it as an option for the buyer's need. We record these separately, including the exact question, platform/interface, date, language or location where visible, and repeated runs where agreed.

03 / Competitors

Who is shortlisted instead?

We look at which businesses appear alongside yours and whether the answer describes a relevant service, specialty or qualification. A competitor's presence is an observation—not proof that a particular page or backlink caused the selection.

04 / Source review

AI citations and supporting pages

When a response exposes citations, we open the URLs, check what they say and record whether the sources are accessible, current and genuinely relevant. We also distinguish an answer citing your page from an answer recommending your company.

05 / Site foundations

Technical SEO and content clarity

We investigate verifiable search blockers such as inaccessible service pages, inconsistent canonical URLs, incorrect business information and thin answers to important buyer questions. We also check whether useful content is discoverable through internal links. These are ordinary web fundamentals; there is no special HTML tag that guarantees AI citations.

Google's published guidance for AI Overviews and AI Mode says that traditional SEO eligibility and helpful, accessible content remain central. See Google Search Central's AI-features documentation and our implementation services.

How the evidence is interpreted

An illustrative audit example

The following is a teaching example, not a completed client audit or a claim about current AI answers.

Illustrative entries in a manual visibility audit
Research questionWhat gets recordedWhat is investigated next
“Which payroll providers support remote teams?”Names in the sampled answer, whether each is offered as an option, and citations if displayedFit of each vendor's service pages, technical availability and the substance of sources
“What should I compare before hiring a payroll provider?”Buyer criteria discussed, sources used and whether any providers are recommendedWhether the business publishes useful, accurate comparison criteria
“Does Provider A offer international payroll?”Specific answer claims and displayed supporting linksWhether official service details support the claims and are up to date

Important: These hypothetical questions and investigation fields are illustrative. They are not evidence that a particular brand appeared or was absent.

What is an appearance rate?

For a defined prompt panel, sampled mention rate = answers mentioning the brand ÷ total sampled answers × 100. If a hypothetical panel contained 12 answers and 3 mentioned the brand, the rate for that panel would be 25%.

The denominator must be shown. This is not a population-wide market share or a permanent platform ranking.

How are citation gaps interpreted?

We check whether a URL appears, whether the linked page supports the relevant claim and whether competitors' cited pages answer the buyer's question more clearly. A missing citation may justify further investigation but does not by itself prove the cause of a recommendation.

For a deeper explanation see how to measure AI visibility.

Our method

How a human-led AI visibility audit is performed

A good audit should be reproducible and understandable to the people making the business decisions.

  1. Agree scope and the buyer journey

    Define your service categories, market, competitors, platforms and what a useful recommendation would mean to your customers.

  2. Choose representative questions

    Combine direct buyer research, keyword/question evidence and product-specific questions. Keep branded and unbranded prompts separate.

  3. Run and log observations

    Use the agreed interfaces and repeat-test design. Record dates, prompts, answers, mentions, recommendations and citation links.

  4. Read the cited sources

    Open important URLs, verify content, investigate competitors and identify mismatches between claims and published information.

  5. Inspect search and technical foundations

    Assess indexing eligibility, page clarity, navigation, metadata and technical blockers with the actual website evidence available.

  6. Prioritize fixes and design follow-up tests

    Report confirmed defects separately from plausible hypotheses, assign practical priorities and explain how to test whether changes matter.

What you receive

An audit report built for decisions, not a decorative score

The final package is agreed in advance. The following describes the proposed report structure, not an existing completed case study.

Evidence appendix

What we observed

A scoped question list, repeat-test conditions, brand and competitor appearance records, and linked sources reviewed where available.

Diagnosis

What the findings support

An explanation of recurring patterns, verified website defects, missing information, content gaps and findings that remain uncertain.

Roadmap

What to do next

A prioritized action plan with suggested owners, effort and confidence, including what should be rechecked after implementation.

Example deliverable structure

Executive brief → scope and core observations → buyer-prompt matrix → platform/competitor observations → source/citation evidence → technical and content findings → 30/60/90-day recommendations → repeat-test plan.

We do not sell an invented universal ranking score or suggest that a screenshot reveals a model's private ranking logic.

Need an audit of how AI sees your business?

We can discuss the buyer questions, markets and platforms that would make the investigation useful before defining deliverables.

Request a manual audit
Scope & pricing

How much does an AI visibility audit cost?

Pricing depends on the research scope. A small business checking one market with a limited question panel needs less work than a multi-region company comparing multiple services, interfaces and competitors. We do not advertise a fixed fee until a deliverable-based package is defined.

The main cost drivers are the number of questions, requested interfaces, runs per question, competitor depth, citation/source review and technical SEO analysis. A concise baseline may be sufficient for an initial decision; a larger business may benefit from repeated testing and ongoing AI visibility monitoring.

Do you need a free check first?

You can run a small manual test yourself using a spreadsheet and publicly available interfaces. That can identify questions worth investigating, but it should not be presented as a statistically representative measurement of the entire market.

How to check AI visibility yourself

Real buyer questions

Manual AI visibility audit: questions and answers

These questions reflect common informational and commercial intents in our keyword research. Related wordings are grouped so each answer stays useful.

What is an AI visibility audit?

An AI visibility audit is a documented examination of how a company appears in sampled AI-assisted search answers. It uses relevant buyer questions to evaluate mentions, recommendations, competitors and displayed citations, then connects verified findings to a prioritized improvement plan.

How do I check the AI visibility of my brand or website?

Begin with buyer questions that describe your services without naming your company. Test them in relevant AI search interfaces, record the exact answers and links, repeat the tests and inspect your site’s crawlability. Our practical guide explains the steps without requiring a paid tool. See the step-by-step guide.

Can I run a free AI visibility audit?

Yes. You can manually test a small, clearly documented question set using interfaces available to you and a spreadsheet. The limitation is that a handful of answers cannot represent every user, query or geographic market. A paid audit adds research design, source review, technical diagnosis and a documented action plan.

What does a manual AI visibility audit include?

A typical scope covers buyer-question research, test conditions, brand and competitor observations, source/citation verification, a targeted technical and content review, and prioritized recommendations. Specific platforms, markets, repeats and report artifacts are agreed before the work begins.

What is a generative engine optimization (GEO) audit?

A GEO audit examines observable brand presence and source evidence in generative search experiences, alongside the web content and technical foundations relevant to discoverability. It is not an examination of a proprietary model’s hidden ranking system.

Is an AEO audit different from a GEO audit?

They overlap. An AEO review emphasizes whether content clearly answers customers’ questions; a GEO review emphasizes how a brand and supporting sources appear in generative answers. Both rely on useful content, search accessibility and evidence-driven assessment.

Can ChatGPT perform an SEO audit by itself?

ChatGPT may help organize a checklist or summarize information supplied to it. That alone is not proof of live crawl access, correct indexing, real prompt coverage, or verified recommendations. Technical findings should be checked against actual website responses and source records.

Which AI platforms can be audited?

Depending on the agreed scope and available capabilities, research may include ChatGPT Search, Google AI Mode, Gemini, Perplexity, Claude and Microsoft Copilot. Results and citation features differ across interfaces and can change over time.

How much does an AI visibility audit cost?

There is no universal fee. Cost is driven by the number of buyer questions, geographic markets, interfaces, repeated runs, competitors, source checks and technical depth. MentionGauge provides a scoped quote before starting; no fixed price is represented on this page.

How often should I audit AI search visibility?

Establish a baseline first. Recheck after important site or business-information changes, and at an agreed regular interval if monitoring is valuable. Use a stable core question panel so comparisons do not silently change the questions being measured.

Does an AI visibility score prove market share?

No. A score calculated from an audit panel represents only the documented test set. For example, three appearances in twelve hypothetical answers would be 25% of that sample—not 25% of all AI searches or prospective buyers.

Can this audit guarantee a recommendation in ChatGPT or AI Overviews?

No. The systems and interfaces are outside our control. An audit identifies observed gaps and defensible changes; it cannot guarantee inclusion, citations, clicks or leads.

Is the audit suitable for a small business or B2B company?

Yes, if the service, market and buyer questions can be defined meaningfully. A local provider might need city-and-service comparisons; a B2B team may need category, integration or vendor-shortlist questions. Scope should fit actual commercial intent.

What happens after the audit?

You can use the report internally, commission targeted AI search optimization or repeat the research to monitor comparable observations. The recommended order usually starts with verified defects and clear business facts before speculative experiments.

Further reading

Source guidance and related research

These are provider documentation links for claims about crawling, indexing and AI search reporting—not endorsements of MentionGauge.

Turn evidence into action

Know what AI answers actually show about your brand.

Discuss a scoped manual AI visibility audit built around your market, your competitors and the questions your buyers ask.