GEO Scoring Methodology

An open methodology for evaluating how well a website is optimized for AI search engines (Generative Engine Optimization).

This scoring system is used by GEOScore to assess website readiness for AI-powered search engines like ChatGPT, Perplexity, Google AI Overviews, and Gemini.

Online Tools

See the methodology in action with these free tools — no installation needed:

• Full GEO Scan — Run all 11 checks automatically and get your GEO Score
• AI Robots.txt Generator — Generate an optimized robots.txt for AI crawlers
• AI Crawler Access Checker — Check if AI search engines can access your site

Why This Matters

AI search engines are fundamentally changing how people discover information. Unlike traditional search engines that rank pages in a list, AI engines:

• Synthesize answers from multiple sources
• Cite specific websites as references
• Evaluate authority and structure differently than Google's PageRank

Websites that aren't optimized for AI search risk becoming invisible to a growing share of search traffic.

The 11 GEO Checks

Our methodology evaluates websites across 11 technical dimensions:

1. AI Crawl Access (Weight: 15%)

What it checks: Whether major AI search engine crawlers can access your content.

Crawler	Operator	Product
GPTBot	OpenAI	ChatGPT
OAI-SearchBot	OpenAI	ChatGPT Search
PerplexityBot	Perplexity	Perplexity AI
Google-Extended	Google	Gemini / AI Overviews
ClaudeBot	Anthropic	Claude
Applebot-Extended	Apple	Apple Intelligence

Scoring:

• All major crawlers allowed: 100%
• Some crawlers blocked: Proportional deduction
• All crawlers blocked: 0%

2. Structured Data (Weight: 12%)

What it checks: JSON-LD / Schema.org markup quality and completeness.

Key schemas evaluated:

• Organization or WebSite on homepage
• Article / BlogPosting on content pages
• FAQPage for Q&A content
• Product for e-commerce
• BreadcrumbList for navigation

Scoring:

• Valid, comprehensive schema: 100%
• Basic schema present: 50-80%
• No schema or errors: 0-30%

3. llms.txt (Weight: 8%)

What it checks: Presence and quality of the llms.txt file.

Criteria:

• File exists at /llms.txt
• Contains site description
• Lists key pages with URLs
• URLs are accessible
• Content is well-structured

Scoring:

• Complete, well-structured file: 100%
• Basic file present: 50-70%
• No file: 0%

Learn more: Complete llms.txt Guide

4. Content Structure (Weight: 12%)

What it checks: How well content is organized for AI extraction.

Criteria:

• Clear H1-H6 heading hierarchy
• Short, focused paragraphs
• Descriptive headings (not ambiguous)
• Lists and tables for structured information
• Key facts stated clearly and early
• FAQ sections with Q&A format

5. Meta Tags (Weight: 8%)

What it checks: Quality of meta information.

Criteria:

• Title tag (unique, descriptive, <60 chars)
• Meta description (compelling, <160 chars)
• Open Graph tags
• Twitter Card tags
• Canonical URL
• Language/hreflang tags

6. HTTP Headers (Weight: 5%)

What it checks: Server response headers.

Criteria:

• HTTPS enabled
• Proper content-type
• Caching headers
• Security headers (HSTS, CSP, etc.)
• Fast response time

7. Sitemap (Weight: 8%)

What it checks: XML sitemap accessibility and completeness.

Criteria:

• Sitemap exists and is accessible
• Referenced in robots.txt
• All important pages included
• No broken URLs in sitemap
• Proper lastmod dates

8. Robots.txt (Weight: 8%)

What it checks: Robots.txt configuration for AI crawlers.

Criteria:

• File exists and is valid
• Doesn't block AI crawlers
• References sitemap
• Logical allow/disallow rules

Learn more: Robots.txt for AI Crawlers

9. Citation Value (Weight: 10%)

What it checks: How likely AI engines are to cite your content.

Criteria:

• Original data or research
• Expert authorship signals
• Statistical claims with sources
• Clear, quotable statements
• Unique insights not found elsewhere

Learn more: How to Get Cited by ChatGPT

10. Internal Linking (Weight: 7%)

What it checks: Link structure quality.

Criteria:

• Logical internal link structure
• Descriptive anchor text
• Hub-spoke content architecture
• No broken internal links
• Reasonable link depth

11. Content Depth (Weight: 7%)

What it checks: Comprehensive content analysis.

Criteria:

• Word count and depth appropriate for topic
• Topic coverage completeness
• Supporting evidence and examples
• Regular content updates
• E-E-A-T signals (expertise, experience, authority, trust)

Learn more: E-E-A-T for AI Search

Overall Score Calculation

GEO Score = Σ (Check Score × Weight) for all 11 checks

Score Range	Rating	Interpretation
90-100	Excellent	Highly optimized for AI search
70-89	Good	Well-positioned with room for improvement
50-69	Fair	Basic optimization, significant gaps
30-49	Poor	Major issues need addressing
0-29	Critical	Essentially invisible to AI search

Automated Scanning

You can run all 11 checks automatically using GEOScore:

1. Free Scan: Instant results across all checks → geoscoreai.com
2. Pro Report ($29): Deep analysis with AI search simulation and actionable recommendations → Pro Upgrade

Research & References

• GEO: Generative Engine Optimization — Foundational research paper
• What is GEO? — Comprehensive introduction
• AI Search Ranking Factors — What factors matter most
• Awesome GEO — Curated resource list

Contributing

We welcome contributions to improve this methodology. Please open an issue or submit a PR with:

• New research findings on AI search ranking factors
• Suggestions for additional checks or criteria
• Real-world data on what drives AI search citations
• Corrections or updates to existing criteria

Related GEO Resources

Free Tools

• GEOScore AI Scanner — Check your website's AI search visibility across 11 signals
• AI Robots.txt Generator — Generate optimized robots.txt for AI crawlers
• AI Crawler Access Checker — Verify which AI bots can access your site

Open Source Projects

• awesome-geo — Curated list of GEO resources, tools, and guides
• ai-robots-txt-generator — Generate optimized robots.txt for AI crawlers
• geo-checklist — Interactive pre-launch GEO readiness checklist
• ai-crawlers-reference — Complete database of AI search engine crawler user-agents
• geo-badge-generator — Generate badges showing your GEO readiness score
• llms-txt-examples — Real-world llms.txt implementation examples by industry
• geo-config-examples — Ready-to-use AI search optimization configs for popular frameworks
• geo-case-studies — Real-world GEO optimization case studies with before/after data
• ai-search-readiness-framework — 11-signal AI search readiness evaluation framework

License

This methodology is released under the MIT License. You are free to use, modify, and distribute it.

---

Maintained by GEOScore — AI Search Visibility Scanner