How GEOAudit Works

Deep dive into our 15-category analysis methodology.

Scoring Methodology

Each webpage is analyzed across 15 categories with 130+ individual checks. Every check results in one of four statuses:

✓

PASS

Full score

⚠

WARN

Half score

✗

FAIL

Zero score

ℹ

INFO

Not scored

Categories are weighted by importance (3%–10% each) and combined into a final 0–100 score with A+ through F grading.

15 Categories in Detail

🧬

Structured Data

10% weight · Category 1/15

Checks for JSON-LD, Schema.org types, critical entities, schema completeness, speakable markup, and more. Structured data is the #1 way AI agents understand your content.

🏗

Semantic HTML

8% weight · Category 2/15

Validates heading hierarchy, semantic elements (header, nav, main, article, section), text-to-HTML ratio, reading level, and proper document structure.

♿

Accessibility

7% weight · Category 3/15

Evaluates HTML lang, image alt text, ARIA landmarks, skip navigation, keyboard accessibility, and other WCAG criteria that also help AI agents parse content.

🔗

Internal Linking

5% weight · Category 4/15

Analyzes internal link structure, anchor text quality, breadcrumb navigation, related content links, canonical URLs, and pagination signals.

🔍

Meta Discoverability

8% weight · Category 5/15

Checks title tag, meta description, canonical URL, Open Graph tags, Twitter Cards, hreflang, author metadata, and publication dates.

🤖

Machine Readability

8% weight · Category 6/15

Evaluates clean HTML structure, content-to-HTML ratio, non-JS-dependent content, data attributes, API endpoints, XML sitemap, and RSS/Atom feeds.

🏛

Entity Authority

6% weight · Category 7/15

Checks organization and author entity definitions, sameAs links, knowledge graph alignment, brand consistency, and expertise credentials.

📌

Citability

8% weight · Category 8/15

Evaluates clear definitions, concise paragraphs, direct answers, quotable snippets, statistical data, comparison tables, and source attribution.

⚡

Performance

5% weight · Category 9/15

Checks image optimization, lazy loading, font-display, DOM size, inline CSS volume, and other performance factors that affect AI crawling.

🤝

Agent Interactivity

5% weight · Category 10/15

Looks for WebMCP manifest, Action schema, API documentation, OpenAPI spec, deep linking, and other machine-actionable interfaces.

📄

LLM Discovery

8% weight · Category 11/15

Detects /llms.txt, /llms-full.txt, .well-known/ai-plugin.json, and other LLM-specific discovery files and their format validity.

🕷

AI Crawler Access

8% weight · Category 12/15

Validates robots.txt rules for AI bots, crawl delays, sitemap presence, noai meta tags, X-Robots-Tag headers, TDM Protocol, and C2PA credentials.

🛡

E-E-A-T Signals

6% weight · Category 13/15

Checks author bios, author schema, publication/modified dates, editorial policy, source citations, expertise indicators, and trust signals.

📝

Content Quality

5% weight · Category 14/15

Evaluates answer-first content pattern, filler phrase detection, content freshness, unique value, and overall content structure quality.

🖼

Multimodal Readiness

3% weight · Category 15/15

Checks image alt descriptiveness, captions, video/audio transcripts, ImageObject/VideoObject schema, responsive images, and SVG accessibility.

Frequently Asked Questions

How long does a scan take?

A typical scan completes in 5–15 seconds depending on the page size and complexity. The Chrome extension runs all 130+ checks directly in your browser for instant results.

Does GEOAudit support single-page applications (SPAs)?

Yes. The Chrome extension analyzes the fully rendered DOM, so it works with React, Vue, Angular, and other SPA frameworks. The web dashboard fetches the raw HTML, so dynamically rendered content may differ.

What's the difference between PASS, WARN, and FAIL?

PASS means the check is fully satisfied (full score). WARN means partial compliance (half score). FAIL means the check is not met (zero score). INFO checks are informational and don't affect scoring.

How are category weights determined?

Weights reflect each category's impact on AI agent discoverability based on research and real-world testing. Structured data (10%) and meta discoverability (8%) are weighted highest because they have the most direct impact on how AI agents parse content.