What are AI visibility (GEO) checks?
AI visibility checks measure whether ChatGPT, Perplexity, Gemini and Claude can read, mention and cite your website. They cover AI crawler access, llms.txt, structured data, entity signals such as Wikidata, and real brand mentions in AI answers. ZORO runs 55 GEO checks and shows the sources AI assistants trust for your category.
Each check shows its impact on your score — High Medium Low — and how to fix it. Click a check for its step-by-step guide.
- GPTBot / OAI-SearchBot accessHighAllow OpenAI's crawlers in robots.txt unless you deliberately want to block them.
- ClaudeBot / Claude-SearchBot accessHighAllow Anthropic's crawlers in robots.txt so Claude can read and cite your pages.
- PerplexityBot accessHighAllow PerplexityBot in robots.txt and in your firewall rules.
- CDN / WAF blocking AI crawlersHighCheck your CDN or firewall bot rules aren't silently blocking AI user agents.
- Content readable without JavaScriptHighMost AI crawlers don't run JavaScript — server-render your key text.
- Brand mentioned in ChatGPTHighEarn mentions on the sources ChatGPT cites: reviews, listicles, Reddit, Wikipedia.
- Brand mentioned in PerplexityHighGet cited by the authoritative pages Perplexity retrieves for your topics.
- Brand in Google AI OverviewsHighRank in the top 10 and provide concise answer blocks that AI Overviews can pull.
- AI share of voice vs. competitorsHighTrack prompts where competitors are named and you aren't, then create content for those gaps.
- Citation-worthy statisticsHighPublish unique, quotable numbers with dates and methodology.
- Concise answer blocksHighAdd 40–80 word direct answers under question headings that AI can lift cleanly.
- Entity clarity (who / what / where)HighState plainly what your company is, what it does and who it serves on the homepage and About page.
- Prompt-aligned contentHighResearch the questions people ask AI assistants and answer them directly on your site.
- Reddit / UGC footprintHighBuild an authentic presence in the communities LLMs cite most.
- Inclusion in "best X" listiclesHighPitch for inclusion in high-ranking "best of" articles that AI tools retrieve.
- Wrong facts in AI answersHighAsk AI assistants about your brand; correct wrong facts on your site and on the sources they cite.
- Bing index coverageHighVerify your site in Bing Webmaster Tools — ChatGPT search and Copilot rely on Bing's index.
- Brand + category co-occurrenceHighGet your brand mentioned alongside your category keywords on third-party sites.
- Competitor citation sourcesHighFind which URLs AI cites for competitors and earn placement on them.
- AI Overview eligibility (top-10 rank)HighImprove rankings for target queries — AI Overviews mostly cite top-ranking pages.
- Google-Extended policyMediumDecide whether Gemini may use your content and set Google-Extended explicitly.
- llms.txt fileMediumPublish /llms.txt listing your key pages with one-line summaries for AI tools.
- Brand in Gemini answersMediumStrengthen entity signals: Knowledge Graph, schema and reviews.
- Brand in Claude answersMediumPublish clear, factual, well-structured pages and earn third-party mentions.
- Brand in Microsoft CopilotMediumOptimise for Bing: Bing Webmaster Tools, IndexNow and Bing Places.
- AI sentiment about your brandMediumFix negative narratives at the source: reviews, forum threads, outdated articles.
- Consistent entity facts across the webMediumUse the same name, description and founding facts on every profile.
- sameAs links in schemaMediumAdd sameAs to Organization schema pointing to LinkedIn, Wikidata, Crunchbase and socials.
- FAQPage schemaMediumMark up genuine FAQs with FAQPage JSON-LD.
- Comparison tablesMediumAdd HTML tables comparing options, prices and features — structured data LLMs extract easily.
- "What is" definitionsMediumWrite clear one-paragraph definitions of the core terms in your niche.
- Author entities and credentialsMediumUse Person schema for authors with credentials and sameAs links.
- Freshness signals for AIMediumShow dateModified in schema and on the page; update key pages regularly.
- Wikidata entityMediumCreate or improve a Wikidata item with accurate facts and references.
- Review platform presence for AIMediumKeep strong profiles on the review sites LLMs frequently cite in your space.
- Pricing information accessibleMediumPublish clear, crawlable pricing so AI answers quote you accurately.
- Specs in HTML, not imagesMediumPut product and service specs in text and tables, not only images or PDFs.
- Question-style headingsMediumPhrase H2s as the natural questions users actually ask.
- TL;DR / key takeawaysMediumAdd a short key-takeaways box at the top of long content.
- Expert quotes and unique POVMediumInclude named expert quotes and clear opinions AI can attribute to you.
- AI referral traffic trackingMediumCreate analytics segments for referrals from chatgpt.com, perplexity.ai and gemini.google.com.
- Passage-level content chunkingMediumWrite self-contained sections that still make sense when extracted alone.
- Local entity data for AIMediumKeep Google, Apple and Bing listings complete — AI assistants use them for local answers.
- Q&A on product pagesMediumAdd customer Q&A sections that mirror conversational queries.
- Knowledge Graph entityMediumConfirm a Knowledge Graph entry exists; strengthen it with schema and Wikidata.
- AI crawler errors in logsMediumCheck server logs for AI bots receiving 403 or 5xx responses.
- Conversational long-tail coverageMediumCreate content for long, specific prompts like "best X for Y under $Z".
- Brand narrative consistencyMediumAlign taglines and positioning across your site, socials and profiles.
- Step-by-step formattingLowFormat processes as numbered steps that language models can summarise.
- Semantic HTML structureLowUse article, section, nav and header tags so parsers understand your layout.
- Context for images and videoLowPair images and videos with descriptive captions and transcripts.
- Outbound citations in contentLowLink to primary sources; well-referenced pages are favoured by LLMs.
- AI usage policy declaredLowState your content usage terms and AI policy explicitly.
- Factual meta descriptionsLowKeep meta descriptions factual and summary-like — some answer engines reuse them.
- Key content only in PDFsLowPublish HTML versions of important PDFs.
AI visibility checks (GEO)
Beyond blue links
These checks measure whether AI assistants can read your site, understand your brand and recommend you: crawler access, llms.txt, structured entity data, answer-friendly content, citation sources, AI Overviews and live mention tests in ChatGPT, Perplexity, Gemini and Claude.
What to fix first
Unblock AI search crawlers, add Organization or LocalBusiness schema with sameAs links, then work on presence on the sources AI cites most in your category.
Frequently asked questions
What is a GEO audit?
A generative engine optimization audit checks whether AI assistants can access, understand and recommend your brand: AI crawler access, llms.txt, entity data, answer-ready content, citation sources and actual mentions in ChatGPT, Perplexity, Gemini and Claude.
Which AI crawlers should I allow?
At minimum the search crawlers that power AI answers — OAI-SearchBot, Claude-SearchBot and PerplexityBot. Training crawlers such as GPTBot, ClaudeBot and Google-Extended are a separate decision.
How do I become a cited source in Google AI Overviews?
Rank well for the query, answer it directly near the top of the page, structure content with clear headings and lists, and keep facts accurate and current.
Does Wikipedia or Wikidata matter for AI visibility?
Both are heavily used sources for entity information. A Wikidata item that lists your official website helps AI systems connect your brand to your domain — if you meet their notability rules.
Check your site for all 55 GEO / AI issues
Run a free audit in minutes: every SEO, AI-visibility, accessibility and security issue, ranked by impact, with the fix for your platform.