How Does Scrutor Calculate Your 0–100 AI Search Visibility Score?
Scrutor provides a deterministic, 100-point Generative Engine Optimization (GEO) audit service for local businesses and online e-commerce brands. Rather than relying on subjective AI guesses, our telemetry engine inspects your live HTTP response headers, /robots.txt AST rules, JSON-LD @graph schemas, and Princeton KDD 2024 citation density across three sequential gates.
Why Is the 3-Gate AI Crawler Stack Sequential?
Retrieval-Augmented Generation (RAG) pipelines operate as a strict funnel. A failure at an upstream gate blocks downstream citation.
Crawler Ingestion & SSR
Verifies that live retrieval bots (OAI-SearchBot, PerplexityBot, Googlebot, Claude-SearchBot) receive HTTP 200 SSR HTML without WAF 403 blocks or client-side JS walls.
Entity Graph & Consensus
Audits machine-readable JSON-LD @graph entities, sameAs authority disambiguation URIs, dateModified velocity, and cross-directory NAP consensus.
RAG Synthesis & KDD Signals
Measures live LLM citation share, 60-word Answer-First architecture, semantic <table> density, and Princeton KDD 2024 multipliers (+41% quotes, +37.2% stats).
What Signals Are Evaluated Inside Each Gate?
Scrutor combines deterministic edge telemetry with live LLM retrieval grounding. Exact sub-signal weight curves and calibration thresholds are proprietary to the Scrutor Neural Index Engine.
| Gate / Pillar | Diagnostic Signal Family | Pillar Allocation | What Our Telemetry Engine Verifies |
|---|---|---|---|
| Gate 1: ACCESS | AI Crawler WAF & RFC 9309 Telemetry | 30% of Total Score | Tests whether edge firewalls (Cloudflare, Sucuri, Wordfence) and /robots.txt rules permit live retrieval agents (OAI-SearchBot, PerplexityBot, Googlebot, Claude-SearchBot), and verifies fast SSR HTML delivery without client-side JavaScript walls. |
| Gate 2: INDEX | Machine-Readable Entity Graph & Consensus | 35% of Total Score | Inspects nested JSON-LD @graph declarations (LocalBusiness, OnlineStore, FAQPage), external knowledge-graph disambiguation links, content freshness signals, and cross-directory entity alignment. |
| Gate 3: CITE | Live RAG Synthesis & KDD 2024 Authority | 35% of Total Score | Benchmarks live brand citation share across ChatGPT Search, Perplexity, and Gemini alongside Princeton KDD 2024 citation drivers (quantitative proof density, attributed quotations, outbound authority links, and answer-first structure). |
Which Peer-Reviewed Standards Power This Methodology?
"Traditional SEO techniques such as keyword stuffing perform poorly in Generative Engines, reducing visibility by 10%. Instead, adding quantitative statistics (+37.2%), quotations from credible sources (+41.0%), and inline citations (+30% to +115%) significantly improves source visibility in synthesized responses."
GEO: Generative Engine Optimization
Princeton University & Allen Institute benchmark establishing Position-Adjusted Word Count (PAWC) citation multipliers.
RFC 9309: Robots Exclusion Protocol
Formal specification for product token matching, longest-match rule precedence, and HTTP 4xx/5xx crawler fallback behavior.
JSON-LD @graph & Entity Linking
Structured vocabulary for LocalBusiness, OnlineStore, FAQPage, and sameAs Knowledge Graph entity reconciliation.
Methodology Frequently Asked Questions
How is Scrutor's 0–100 AI Visibility Score structured?
Scrutor evaluates domains across three sequential pillars: Gate 1 Access (30% weight), Gate 2 Index (35% weight), and Gate 3 Cite (35% weight). Over 25 proprietary telemetry signals—spanning live edge crawler probes, RFC 9309 directives, JSON-LD knowledge graph validation, and academic citation linguistics—are combined into a single 0–100 readiness index.
Why does Scrutor evaluate a Dual-Layer crawler policy?
AI crawlers operate in two distinct roles: foundation-model training scrapers and real-time search retrieval agents (such as OAI-SearchBot, PerplexityBot, Claude-SearchBot, and Googlebot). Our engine verifies that your security perimeter and exclusion directives permit live answer-engine grounding without exposing proprietary assets unnecessarily.
What academic research informs Gate 3 (Citation Authority)?
Gate 3 incorporates empirical findings from the Princeton University, Georgia Tech, and Allen Institute for AI study on Generative Engine Optimization (ACM SIGKDD 2024, arXiv:2311.09735), which demonstrated that verifiable quantitative claims, attributed quotations, and authoritative citations significantly outperform legacy keyword-density tactics in RAG synthesis.
Why do client-side rendered (CSR) websites struggle in AI search?
Unlike traditional search indexers that queue pages for deferred headless browser rendering, real-time AI answer engines fetch live pages under strict millisecond latency budgets. If core entity facts are trapped inside unhydrated JavaScript bundles, retrieval agents fail to extract them.
Test Your Domain Against All 100 Points in 30 Seconds
Run a free 3-Gate diagnostic right now, or preview a complete 8-page Executive Engineering PDF Dossier before you audit.