Quick answer: An AI search visibility audit checks whether AI tools such as ChatGPT, Perplexity, Gemini and Google's AI Overviews can find, understand, trust and cite your brand. It covers five areas: access (can AI crawlers and search indexes reach your pages?), content (are your answers clear and quotable?), entity (do AI tools know who you are and describe you consistently?), authority (do trusted sources mention you?) and measurement (are you tracking what AI tools actually say?). Use the 30-point checklist below to find the gaps, then fix them in that order.
How to use this checklist
Work through each section and mark every item as done, partly done or missing. Fix access problems first, because nothing else matters if AI tools can't reach your pages. Then improve content and entity signals, build authority and set up measurement so you can see what changes.
1. Access: can AI tools reach your content?
- Key pages are indexed in Google. Check in Search Console's URL Inspection. AI Overviews and AI Mode only cite pages that are indexed and eligible for a snippet.
- Key pages are indexed in Bing. Verify the site in Bing Webmaster Tools. Several AI assistants draw on Bing's index, and ChatGPT search launched relying largely on it.
- robots.txt allows the AI crawlers you want. For example, OpenAI documents OAI-SearchBot as the crawler for ChatGPT search, separate from GPTBot, which is used for model training. You can allow one and block the other.
- No accidental blocks from firewalls, bot protection or CDN rules that stop legitimate crawlers.
- Important content is in the HTML, not only rendered by scripts or hidden behind logins, tabs that load on click, or PDFs alone.
- No snippet restrictions (
nosnippet, very lowmax-snippet) on pages you want quoted. - XML sitemap is current and submitted to Google and Bing. Consider IndexNow so Bing and other engines hear about new or updated pages straight away.
2. Content: are your answers easy to quote?
- Each key page answers its main question in the first two or three sentences.
- Headings use the questions buyers actually ask, taken from Search Console, sales calls and support tickets.
- Definitions are plain and self-contained: "X is Y that does Z".
- Specific facts and numbers instead of vague claims, with sources cited.
- Comparisons and steps are in tables and lists.
- FAQ sections cover the follow-up questions: cost, timing, alternatives, risks, how to choose.
- Bottom-of-funnel pages exist: comparisons, alternatives, pricing and use cases, which AI tools draw on for recommendations.
- Content is current, with visible "last updated" dates on pages that change.
3. Entity: do AI tools know who you are?
- Organization structured data on the site: name, logo, address, contact details and
sameAslinks to your official profiles. - Consistent name and description everywhere: website, LinkedIn, Crunchbase, Clutch, G2, Google Business Profile and directories.
- Former names are connected. If you've rebranded, use
alternateNamein your structured data and update old profiles, so AI tools don't treat you as two different companies. - A clear About page stating what you do, for whom, where, since when and who leads the company.
- Author and expert pages with credentials, linked from their content and to their external profiles.
- Service, product and article structured data that matches what is visible on each page.
4. Authority: do trusted sources mention you?
- Listed on the review sites and directories that matter in your category, with complete profiles and recent reviews.
- Mentioned in industry publications through earned coverage, guest articles, interviews or podcasts.
- Included in third-party "best of" lists and comparisons where it is legitimate to be.
- Original data or research others can cite.
- Active, accurate presence in communities your buyers use, such as relevant subreddits, forums and LinkedIn, where AI tools increasingly look for real opinions.
5. Measurement: what do AI tools actually say?
- A fixed list of 20–50 buyer questions covering your category, your competitors and "best X for Y" queries.
- Monthly checks across ChatGPT, Perplexity, Gemini and Google AI Overviews, recording whether you are mentioned, cited and described correctly, and who else appears.
- Referral traffic from AI tools tracked in analytics (chatgpt.com, perplexity.ai, gemini.google.com and others).
- A log of inaccuracies in how AI tools describe you, with the source pages that probably caused them, so you can fix them at source.
From our own audit of ballistic.media: running this checklist on our own site in November 2026 found three problems we then fixed:
- 43 of 117 old blog posts were "Crawled – currently not indexed" by Google, so no AI tool relying on Google could cite them. We merged or removed weak posts so the strongest ones carry the topic.
- Our former name, Ballistic Design Studio, wasn't connected to the current brand. We added it as an
alternateNamein our Organization structured data. - Bing heard about new content slowly. We added IndexNow so new and updated pages are submitted the moment they are published.
We also found that ballistic.media already appeared for 1,217 long, chatbot-style questions in Google Search in 90 days, a sign of how much research AI assistants already do on buyers' behalf.
What about llms.txt?
llms.txt is a proposed file, published in September 2024, that gives AI tools a curated map of a site's content. It is cheap to add, and some documentation tools and coding agents use it, but it is not a ratified standard. Google has said it does not use it for Search. Treat it as optional: fixing access, content and entity signals matters far more.
Where Ballistic Media fits: we run AI search visibility audits for B2B, SaaS and fintech brands: all 30 checks plus a baseline of how ChatGPT, Perplexity, Gemini and AI Overviews describe you today, and a prioritised plan. See our AI search optimisation services, or read AEO vs GEO vs SEO and how to measure AI search visibility.
FAQs
How long does an AI search visibility audit take?
A first pass through the checklist takes a day or two for a small site. Building a proper baseline of how AI tools describe you, across 20–50 questions and four or more tools, adds a few more days.
Should we block AI crawlers?
It depends on your goals. Blocking training crawlers like GPTBot keeps your content out of model training, while allowing search crawlers like OAI-SearchBot keeps you eligible to appear in AI search answers. Many companies allow search crawlers and decide separately about training.
How often should we re-run the audit?
Run the full checklist every six months, and the measurement section (questions across AI tools) monthly. AI answers change often, so trends over several months matter more than any single check.
Can we fix how ChatGPT describes our company?
Not directly. You can correct the sources it draws on: your own site, your profiles and directories, and third-party articles. Clear, consistent and well-cited information across those sources is what changes the answers over time.
Is AI search visibility different from SEO?
It builds on SEO. Most of the access and content checks are SEO fundamentals. The entity, authority and measurement sections reflect what matters more in AI answers: how consistently and how often trusted sources describe you.
Sources
- Google Search Central, AI features and your website
- OpenAI, Overview of OpenAI crawlers
- Search Engine Roundtable, ChatGPT Search is powered by Bing's index and more
- llmstxt.org, the llms.txt proposal
- IndexNow
- Ballistic Media site audit and Search Console data, November 2026




