Media
3 items
Description
Rule-based audit of how well a site is structured for AI search to understand and cite. Runs locally: no AI, no backend. GEO Auditor scores how well a website is structured for AI search systems to understand, extract, and cite — and shows you exactly which observable signals produced the score. Every point is traceable to a named rule. Nothing is guessed, and no AI model is involved. WHAT IT CHECKS • Entity Clarity (20%) — Can the site be resolved to a named entity with a stated category, audience, and use case? • Answerability (25%) — Does the site directly answer high-intent questions: what is this, who is it for, how does it work, what does it cost, how does it compare? • Evidence (15%) — Do statistical claims carry a nearby source, study, or methodology? Are authors and dates present? • Authority Signals (15%) — Case studies, named testimonials, certifications, team information, trust pages. • Content Structure (10%) — Headings, section segmentation, lists and tables, descriptive anchors, semantic HTML. • Technical AI Accessibility (10%) — Indexability, server-rendered content, robots.txt, sitemap, canonicals, valid structured data. • Freshness (5%) — Whether date signals are present at all. HOW IT WORKS 1. Open any website and click the toolbar icon for an instant audit of the current page, using the rendered DOM so single-page-app content is included. 2. Click "Run full site audit" to crawl the site — robots.txt, sitemap.xml, and priority-ordered link traversal — with live progress and working Pause and Stop controls. 3. Read the full report: score breakdown, top opportunities ranked by score impact, question coverage, per-claim evidence analysis, technical accessibility, and page-by-page results. 4. Export as JSON, CSV, or a standalone client-ready HTML report. Everything is generated in your browser. EVERY SCORE IS EXPLAINED Expand any finding to see three things: what we detected, why it matters, and how to fix it — plus the rule ID, weight, and severity. No opaque numbers. BUILT FOR SEO CONSULTANTS, AGENCIES, AND IN-HOUSE TEAMS The exported HTML report is presentable to a client as-is: executive summary, score breakdown, and a prioritised list of opportunities. SEO AND GEO ARE KEPT SEPARATE Classic SEO hygiene (title, H1, canonical, robots, sitemap, schema) gets its own Technical Foundation section and is deliberately capped at 10% of the score. This is not another SEO checker. PRIVACY BY DESIGN • All analysis runs locally in your browser. • No backend, no account, no login, no analytics, no tracking. • No page content is transmitted anywhere. The only network requests are direct fetches of the pages you are auditing, plus that site's robots.txt and sitemap.xml. • No AI model or external API is called. DETERMINISTIC AND REPRODUCIBLE Scan the same unchanged site twice and you get the same score. 65 rules, fixed weights held in a single configuration, no randomness. WHAT THIS SCORE IS NOT This score evaluates observable website signals that can affect how clearly a website's information can be understood and retrieved by AI and search systems. It does not predict rankings in ChatGPT, Gemini, Perplexity, or Google AI, and it cannot verify whether any claim on a site is true — only whether supporting evidence is present. KNOWN LIMITATIONS • Pattern detection is tuned for English. Non-English sites are scored structurally, but language-dependent rules will under-report. • Pages beyond the starting page are fetched as raw HTML rather than executed, which reflects what a non-JavaScript crawler would see. • Authority signals are observable declarations only — a certification or award cannot be independently verified by an extension.