GEOAuditChecklist
Measure your website's technical readiness for the age of AI Search.
Readiness Score
Based on weighted technical factors including bot access, semantic structure, and data schema.
1. Technical Access & Bot Governance
Ensures AI crawlers can physically access and read your content without being blocked.
2. LLM Standards & Manifests
New protocols specifically designed to help Large Language Models navigate your site.
3. Semantic Structure & HTML5
How your code helps the AI understand what is a heading, what is data, and what is junk.
4. Entity Knowledge Graph (Schema)
Structured data that explicitly tells the AI 'Who, What, Where' without ambiguity.
5. Trust, Authority & E-E-A-T
Factors that reduce the 'Hallucination Risk' score of your content.
6. Content Optimization (GEO)
Writing styles that increase the likelihood of being cited as the source.
Methodology
How AEO scoring actually works
The weighting is not arbitrary. It mirrors what AI engines do at retrieval time.
Block citation entirely if missing
JavaScript-only rendering, missing canonical claims, and content that fails GIST Granularity will get a site excluded from AI citations regardless of how strong everything else is. These items are the gating layer.
Reduce citation probability significantly
Missing schema, undated articles, no author bios, weak internal linking. Each one is recoverable, but their accumulation is what separates strong sites from invisible ones in AI search results.
Marginal but compounding
BreadcrumbList schema, link text quality, alt text completeness. None of these alone moves the needle, but a site that handles all three signals editorial care to AI engines.
What each tier means
Under 40%: structural problems blocking citation. 40-60%: technical baseline solid, content quality the limiting factor. 60-80%: above most AI search competitors. Over 80%: positioned for consistent citation across major AI engines.
Glossary
Terms used in this audit
Plain-text definitions for the acronyms and concepts referenced throughout.
- AEO
- Answer Engine Optimization. Optimizing content for direct-answer AI engines like ChatGPT, Perplexity, and Claude. AEO rewards atomic, citable claims with clear attribution. Where SEO asks "does the page rank?", AEO asks "does this passage get pulled into the answer?"
- GEO
- Generative Engine Optimization. Optimizing for AI-powered search experiences like Google AI Overviews and SGE. GEO rewards structured, factual passages that AI engines can extract and cite alongside their generated summaries.
- GIST
- Granularity, Inferability, Specificity, Topicality. The four content properties that determine whether AI engines will cite a passage. Granular claims (atomic, single-sentence units), inferable without prior context, specific in their assertions (named tools, numbers, versions), and topically aligned with their headings.
- llms.txt
- A proposed AI-discovery file. Lives at the root of a domain (yoursite.com/llms.txt) and tells AI assistants what the site is, what content matters most, and how to navigate it. Proposed by Jeremy Howard in September 2024. By mid-2026, adoption sits around 10% of domains; SEO citation impact is low, but IDE agents and MCP servers route on it routinely.
- E-E-A-T
- Experience, Expertise, Authoritativeness, Trustworthiness. Google's quality framework, now mirrored by every major AI engine. Author identity, source attribution, original research, and editorial signals all feed into whether AI engines treat a site as cite-worthy.
- AI Overviews
- Google's AI-generated answer summaries. Appear above traditional search results for many queries. Sources cited in AI Overviews receive a share of impressions and clicks. Inclusion correlates with structured content, schema markup, and high topical authority.
- JSON-LD
- JavaScript Object Notation for Linked Data. The structured-data format Google explicitly prefers for schema markup. Lives in a single <script type="application/ld+json"> block in the page head. AI engines parse JSON-LD more reliably than Microdata or RDFa.
FAQ
Questions worth answering
What is the difference between SEO, GEO, and AEO?+
SEO (Search Engine Optimization) optimizes for traditional ranked search results. GEO (Generative Engine Optimization) optimizes for AI-powered search experiences like Google AI Overviews and SGE. AEO (Answer Engine Optimization) optimizes for direct-answer engines like ChatGPT, Perplexity, and Claude. The three overlap but require different content properties: SEO rewards depth and links, GEO rewards structured factual passages, AEO rewards atomic citable claims.
How long does a full audit take?+
About 45 minutes for an experienced SEO going through all 40 items on a single page. Less for a quick spot-check, more for a full site audit across multiple pages. The checklist is designed to be tackled section by section. Your progress saves automatically in your browser, so you can come back to it.
Is this checklist enough to fix everything?+
The checklist identifies what needs fixing. Fixing requires either manual content rewrites or an automated tool. For an automated audit that runs the same checks against any URL and ranks the gaps by impact, use the site audit at /get-started. For asset generation (schema markup, FAQ blocks, meta tags), use the free GEO Asset Generator.
Does my score predict actual AI citation rates?+
Indirectly. The 40 items in this checklist correlate with citation probability, but they do not guarantee citations. AI citation is influenced by query, model, retrieval freshness, and competitive landscape. A high score makes citation possible; a low score makes it unlikely.
How are points weighted?+
Items are worth 1, 2, or 3 points based on impact. Granularity, Specificity, and JS-only rendering are 3-point items because they're the most common failure modes. BreadcrumbList schema is 1 point because it helps but isn't critical. The total max score is 89 points across all 40 items. Your displayed score is the percentage of total points achieved.
Should I run this on every page or just the homepage?+
Both. Some items (llms.txt, robots.txt, Organization schema) are site-wide. Others (BlogPosting schema, paragraph length, factual density) are per-page. Run it once for site-wide items, then pick your top 5 highest-traffic pages and run the per-page items on each.
Report generated via WebsiteAIScore.com