HomeProductsAI SIGNALS™
Product 04 · Signal Layer

AI SIGNALS™
Build Machine-Readable Visibility.

AI SIGNALS™ generates, organises, validates, and maintains the 167 atomic machine-readable signals that AI systems use to discover, classify, verify, and recommend businesses. The structured signal layer is the difference between being on a website and being in the AI knowledge graph.

167Atomic Signals
3Categories
4Validation Stages
LiveMonitoring
The Problem

AI Systems Read Signals. Not Websites.

Most businesses understand that content matters for search engines. Far fewer understand that AI systems do not primarily read content — they read signals. Structured, atomic, machine-readable data points that tell AI systems precisely what a business is, what it does, who it serves, what it has proven, and whether it can be trusted.

These signals are not embedded in website copy. They are deployed in structured files, schema markup, edge-delivered JSON, and linked data graphs. A business without a signal layer is not just less visible — it is categorically different in how AI systems process and weight it.

AI SIGNALS™ builds and maintains the complete signal layer — 167 signals across three categories — ensuring every signal is present, correctly structured, consistently maintained, and accessible to every AI system that queries it.

The 167 Signals

Three Categories. Every Signal Has a Purpose.

Each category targets a distinct AI system behaviour. All three must be complete for full signal coverage.

48

Identity & Entity

Signals that establish who the business is at the entity level. Legal name, all name variants, category, NACE/SIC code, founding date, headquarters, all locations, canonical URL, social profiles, registration number, VAT, entity type, entity status, parent entity, subsidiaries, and all sameAs references to authoritative external profiles.

entity.legal-name entity.category entity.nace-code entity.founding-date entity.same-as[] entity.canonical-url
61

Trust & Proof

Signals that prove the business is who it claims to be. All certification references with issuing body URLs, awards, professional body memberships, case study schema, client testimonials, third-party reviews, citation count, sector directory presence, SHA-256 proof files, OpenTimestamps anchors, and all verification statuses.

proof.sha256 proof.ots-anchor trust.certifications[] trust.citations[] trust.verification-status trust.case-outcomes[]
58

Intent & Governance

Signals that map the business to buyer intent and declare its governance posture. Primary and secondary service categories, all service-to-intent mappings, buyer personas, problem solved, delivery formats, pricing model, geographic and sector intent filters, negative intents, AI crawler permissions, and all governance file statuses.

intent.service-category intent.buyer-persona[] intent.negative[] governance.eu-ai-act governance.gdpr governance.hitl-status
The Canonical Rule

Signals → Files. Never Files → Signals.

The ADI™ Pipeline Rule

Signal values are determined first — from the Business Knowledge Graph™. Files are generated from those signal values. Reversing this order produces inconsistent, error-prone deployments that create conflicting data in the AI knowledge graph. This rule is never violated.

RAW FACT ENTITY RELATIONSHIP SIGNAL FILE
Signal Validation

Four Validation Stages Before Deployment

Every signal passes through four validation stages. No unvalidated signal is ever deployed.

01

Completeness

Is every required property present? Missing required fields are flagged before any file is generated.

02

Consistency

Does this signal conflict with any other signal in the set? Cross-signal conflicts are resolved before deployment.

03

Verifiability

Can this signal's value be confirmed from an external source? Unverifiable signals are flagged for human review.

04

Accessibility

Is this signal accessible to the AI crawlers that need it? Deployment path verified before the signal goes live.

Output Files

Four Structured Deliverables

AI Signals Registry

Complete Signal Manifest

All 167 signals in a versioned registry with current status, last validation date, next review date, and source reference for every signal.

Signal Files

All Structured Signal Files

ai.json · entities.json · intents.json · ai-actions.json · ai-routes.json · ai-validation.json · allow-lane-matrix.json · adn.json — all generated from the signal set.

Signal Validation Report

Status of Every Signal

Structured report of every signal's validation status — completeness, consistency, verifiability, and accessibility — with findings and corrective actions for every failing signal.

Signal Monitoring

Continuous Health Dashboard

Real-time view of signal health across all 167 signals with trend data and anomaly alerts. Powered by AI SONAR™ once the signal layer is deployed.

The Result

Before and After AI SIGNALS™

Before
  • Some schema markup exists — no complete signal layer
  • Signal files incomplete, inconsistent, or absent
  • No signal validation — errors go undetected
  • AI classifies business incorrectly or partially
  • No monitoring — signal decay goes unnoticed
  • Buyer intent not mapped to any service
  • Governance posture not declared — AI trust low
After
  • All 167 signals deployed and validated
  • Eight signal files live and edge-accessible
  • Four-stage validation — no invalid signal deployed
  • AI correctly classifies business across all dimensions
  • Continuous monitoring — signal decay detected immediately
  • Buyer intent mapped — business appears for right queries
  • Governance declared — AI treats business as trustworthy
Next Step

From Signals to Graph

Product 05

KNOWLEDGE GRAPH™

KNOWLEDGE GRAPH™ builds the complete entity relationship graph — connecting every entity, typing every relationship, and publishing the result in JSON-LD for AI knowledge graph integration.

Go to KNOWLEDGE GRAPH™ →
Previous

AI READY™

AI SIGNALS™ builds on the files generated by AI READY™. If not yet complete, start with AI READY™.

← Back to AI READY™