JSON-LD Person Schema: The Most Important Structured Data for AI Visibility — NovCog AI Practitioner Series by Guerin Green

09-person-schema-guide.md

JSON-LD Person Schema: The Most Important Structured Data for AI Visibility

By Guerin Green / Novel Cognition AI Strategist | Federal Courthouse AI Presenter

---

Part of the NovCog AI Practitioner Series

---

Of all schema types, Person markup has the highest correlation with AI citation. The 70.4% finding makes the case statistically, but the implementation details determine whether your markup actually works — or gets ignored.

This guide covers the complete implementation pattern for Person schema in JSON-LD, including the properties most practitioners miss.

The Minimum Viable Schema

Every Person schema deployment needs these core properties:

{
  "@context": "https://schema.org",
  "@type": "Person",
  "name": "Full Legal Name",
  "jobTitle": "Primary Professional Title",
  "description": "One-sentence description of expertise and role",
  "url": "https://canonical-website.com",
  "image": "https://canonical-website.com/headshot.jpg",
  "sameAs": [
    "https://linkedin.com/in/profile",
    "https://twitter.com/handle",
    "https://www.wikidata.org/wiki/Q-number"
  ]
}

This is the floor, not the ceiling. Most practitioners stop here and wonder why their entity isn't resolving in knowledge graphs.

Properties That Move the Needle

knowsAbout — An array of topics the person has expertise in. This directly informs how AI systems categorize the entity:

"knowsAbout": [
  "Retrieval-Augmented Generation",
  "Structured Data Strategy",
  "AI Entity Authority",
  "Search Engine Optimization"
]

performerIn — Links the person to events. This is critical for public figures, speakers, and anyone whose authority derives from appearances. It creates an Event-Person relationship that knowledge graphs traverse:

"performerIn": {
  "@type": "Event",
  "name": "AI and the Law: CLE Presentation",
  "location": {
    "@type": "Place",
    "name": "Federal Courthouse"
  },
  "startDate": "2025-01-15"
}

affiliation — Organizational connections transfer authority bidirectionally. An affiliation with a recognized institution increases the person's entity confidence:

"affiliation": {
  "@type": "Organization",
  "name": "Organization Name",
  "url": "https://organization-website.com"
}

colleague — Declares peer relationships. In a Distributed Authority Network, colleague references across multiple properties create dense entity graph connections:

"colleague": [
  {
    "@type": "Person",
    "name": "Colleague Name",
    "url": "https://colleague-website.com"
  }
]

The sameAs Chain

sameAs is the most powerful single property for entity disambiguation. Each URL in the array tells machines: "This entity on this page is the same entity at that URL."

Priority order for sameAs (by impact on entity resolution):

1. Wikidata — The canonical linked data source. If you can create a legitimate Wikidata entry, this is the strongest single signal.
2. LinkedIn — Professional identity confirmation.
3. Google Scholar / ORCID — Academic authority.
4. Professional organization profiles — Institutional confirmation.
5. Social media profiles — Breadth of presence confirmation.
6. Speaking engagement pages — Event-based authority.
7. DAN properties — Cross-domain entity confirmation within your network.

Eight or more sameAs links is the target. Each one is an independent vector confirming entity identity.

Deployment Across a Network

In a DAN context, every property deploys Person schema for the central entity. The critical rule: consistency across all deployments. The name, jobTitle, description, and sameAs array must be identical across all properties. Inconsistencies cause entity fragmentation — the opposite of what you want.

The properties that vary per deployment:

• url (matches the current property)
• performerIn (relevant events for that property's context)
• colleague (relevant peers for that property's context)
• Additional @type arrays (e.g., ["Person", "Author"] on a publishing site)

See Schema for Public Figures for the expanded pattern used in political and public figure contexts.

Validation and Testing

Before deploying, validate with:

1. Google Rich Results Test — Confirms Google can parse your markup
2. Schema.org Validator — Checks syntax compliance
3. Google Search Console — Monitors structured data status post-deployment
4. Manual entity query — Ask ChatGPT, Claude, or Gemini about the entity. If structured data is working, the AI should return attributes that match your schema.

The markup itself is simple. The strategy — deploying it consistently across multiple properties with dense sameAs chains and network-wide colleague references — is what builds the entity authority that earns AI citation.

For the full terminology reference, see the RAG Glossary.

---

Series Navigation

• A Practitioner's Guide to LLMs — Start here
• Verifying Googlebot
• Person Schema Guide — You are here
• Wayback Recovery
• Entity Authority
• Schema for Public Figures
• Complete Series Index

---

Burstiness & Perplexity Community | Hidden State Drift | Novel Cognition

#hiddenstatedrift #burstinessandperplexity