Can an AI visibility tool tell me why AI ignores my brand?

An AI visibility tool can tell you some of the reasons AI ignores your brand. It cannot tell you all of them, and it cannot read the model's mind.

The reason is how an AI answer gets built. It passes through four stages:

1. Parametric: what the model knows from training, frozen at a cutoff. Nobody changes this quickly.
2. Access: whether a crawler can fetch and store your page. You and your CDN control this.
3. Retrieval: whether your page enters the model's context for a given query. The engine controls this.
4. Generation: which retrieved pages the model uses and cites. The model controls this.

A tool sitting outside the engine can test stage 2 directly, because it can request your pages the way a crawler does. It can also check what your site says about you. Stages 3 and 4 happen inside the engine, so the tool only sees the final answer.

That matters for diagnosis. If a tool finds a block at the access stage, you have a real, fixable cause. If it finds nothing, the remaining reasons sit where nobody outside can look. This is also why a tool that fixes beats one that only tracks: the fixable causes are the ones it can actually see.

These checks have a documented mechanism and a clear pass or fail.

Crawler permission, bot by bot

robots.txt rules are per bot. OpenAI's crawler documentation says GPTBot governs training, OAI-SearchBot governs ChatGPT search, and "each setting is independent of the others." A common error is a blanket AI block that removes you from ChatGPT search when you only meant to opt out of training.

Firewall and CDN blocks

OpenAI's help centre says eligibility means allowing OAI-SearchBot in robots.txt and confirming your host or CDN lets its traffic through. A firewall can return 403 to a crawler that robots.txt allows. Nothing in your dashboards will show it. As of October 2026, note that Cloudflare began blocking Training and Agent crawlers by default on ad-displaying pages from 2026-09-15 for new domains and existing free-tier customers, while leaving Search crawlers allowed.

Rendering gaps

Vercel's crawler study found ChatGPT and Claude crawlers fetch JavaScript files but "don't execute them. They can't read client-side rendered content." If your pricing or product names load only after JavaScript runs, those crawlers never see them. Googlebot renders, so this gap does not apply to Google.

Accidental suppression

noindex, nosnippet, max-snippet:0 and data-nosnippet keep content out of AI answers on Google and Bing. Templates sometimes carry them by mistake.

Contradictory facts

If your pricing page, about page and footer state different numbers, an engine has conflicting sentences to quote. A tool can extract and compare them. Our guide to fixing contradictions in your brand facts covers the cleanup.

Some causes are invisible to every outside tool, including ours.

• The real retrieval queries. OpenAI says ChatGPT search rewrites a prompt into one or more targeted queries. Google may fan a question out across subtopics. A tool that tests your chosen prompts is testing a proxy for an input it cannot observe.
• Position in the context. C-SEO Bench found that a source's position in the model's context matters far more than content tweaks. That position is set inside the engine. No publisher can see or set it.
• Ranking factors. OpenAI's public description is that ChatGPT ranks results using "multiple factors" and that placement is not guaranteed. Nothing more is enumerated.
• Whether search ran at all. Retrieval is per query and discretionary. Any answer may come from training data alone, and you often cannot tell which.

The July 2026 critical survey of 45 GEO studies (arXiv:2607.14035) concluded that no reviewed technique shows a stable, cross-platform causal effect on discoverability. If the research cannot establish the levers, a dashboard cannot either. Google Analytics has the same blind spot, covered in what Google Analytics can and cannot tell you about AI answers.

Watch for these claims, because the evidence does not support them:

• "The model skipped you because of low authority." No engine publishes an authority score. Domain Authority is a Moz metric no engine computes.
• "Add schema and you'll get cited." Google says structured data isn't required for generative AI search. Ahrefs' matched test found AI Overview citations fell 4.6% after pages added schema.
• "Your AI rank is 4." Google's AI reporting exposes impressions only. There is no position data to check such a number against.
• "Improve your E-E-A-T score." Google says E-E-A-T isn't a specific ranking factor. There is no score.

Ask any vendor one question: which of your numbers are measurements, and which are your model? An honest tool can answer. Any single visibility score, ours included, averages answers given with and without live retrieval. Our playbook on explaining score limits to stakeholders helps you frame this internally, and our checklist for switching AI visibility tools turns it into vendor questions.

Work through the fixable causes first, in this order:

1. Check per-bot robots.txt. Confirm OAI-SearchBot, PerplexityBot and Claude-SearchBot are allowed if you want to appear in those engines' search.
2. Test real reachability. Request a page as each AI crawler and look for 403s or challenge pages. The AI Crawler Checker does this and shows which crawlers your firewall turns away.
3. Diff raw HTML against the rendered page. Anything that appears only after JavaScript runs is a candidate for server-side rendering.
4. Search your templates for snippet controls you did not mean to set.
5. Reconcile your facts so one price, one founding year and one product description appear everywhere.

Then measure each engine separately. The free checker on this site shows whether AI names you for up to three buyer questions, as a snapshot. Once access is clean, what the model says is not in your control. Being named on the sites engines already cite is the remaining route, and it carries no guarantee. For how this differs from classic site checks, see an AI visibility audit versus an SEO audit.

If my robots.txt allows AI crawlers, am I definitely reachable? No. robots.txt states permission, but a firewall, CDN or bot-management rule can still return 403 or a challenge page to the crawler. OpenAI's help centre says eligibility for ChatGPT search needs both: OAI-SearchBot allowed and your host or CDN letting its traffic through. Test with real crawler requests, not just by reading the file.

Does blocking GPTBot remove me from ChatGPT search? No. OpenAI documents GPTBot as a training crawler and OAI-SearchBot as the search crawler, and says each setting is independent. You can block GPTBot to opt out of training while allowing OAI-SearchBot so your pages stay eligible for ChatGPT search answers. Allowing GPTBot does not get you cited either.

Can a tool tell me which queries ChatGPT ran to answer about my brand? Not from the outside. ChatGPT rewrites prompts into its own targeted queries and does not expose them on the consumer app. Prompt-tracking tools test questions they chose, which is a proxy. Bing Webmaster Tools' AI Performance report shows grounding queries for Copilot and Bing AI summaries, but only in its interface, as of October 2026.

If every access check passes and AI still ignores me, what then? Then the remaining causes sit in retrieval and generation, which the engine controls. Keep facts specific and consistent on fetchable pages, since a stated price or one-line definition gives the engine a sentence to quote. Being mentioned on the sites engines already cite is associated with appearing in answers. Neither path is a guaranteed outcome.

Why does AI get facts about my brand wrong? Often the model answers from training data frozen at a cutoff, so newer prices or products are missing. Sometimes it retrieves pages that contradict each other, including your own. A tool can show which facts each platform gets wrong and whether your site states them consistently. Fixing your own contradictions is the part fully in your control.