Entity Architecture and Governance Framework (EP-9)

What the EP-9 framework is

The EP-9 framework is a deterministic nine-layer specification for establishing an entity, separating it from namesakes and preparing it for extraction by AI search. It starts from one observation: generative search engines such as Google AI Overviews, ChatGPT, Perplexity and Gemini weigh clear entity recognition, verified Knowledge Graph links and machine-readable schema, not keyword frequency alone.

The nine layers form one sequence, which the specification calls the Trust Chain. Each layer depends on the one before it: a schema declaration has nothing to confirm until the entity has a page, and an external profile has nothing to point to until the entity has a canonical URL.

Trust Chain: Clear Identity (L1) → Owned Content (L2) → Contextual Links (L3) → DefinedTerm Schema (L4) → Disambiguation (L5) → sameAs KG Anchoring (L6) → Proof Stack (L7) → DOM QA (L8) → Off-Site Sync (L9)

What an entity is in EP-9

An entity in EP-9 is more than a name that appears on a page. The specification defines it by four properties at once: an identity, a context, relationships and evidence. The sentence below is the definition as deposited, and it is also the exact text of this page's DefinedTerm schema, as checkpoint CHK-EP-07 requires.

An entity with a clearly defined identity, a context that separates it from namesakes, links to related entities, and enough evidence for search engines, AI systems and people to understand what it is and why it can be trusted.

A keyword helps a page be found. An entity helps a page be understood, trusted and quoted, because a machine can tell what the page is about and connect it to other things it already knows.

Why owned pages come before off-site profiles

Owned pages come before off-site profiles because, in the words of the specification, off-site social profiles and external citations cannot substitute for on-site owned entity architecture. Topical authority needs a foundation on the website itself: dedicated entity pages, structured DefinedTerm schema and deliberate internal links.

A profile created on a social network with no content behind it adds a name, not an entity. Off-site consistency is therefore the ninth layer, not the first. It confirms an entity that the first eight layers have already built.

The four entity functional groups

The four entity functional groups sort every entity on a domain by the job it does for a buyer: Offer, Problem, Method and Trust. Grouping entities this way prevents topical clutter and ties each one to a stage of Decision Ladder Navigation (DLN), so each entity page knows which next step it should offer.

The nine layers and their priority

The nine layers of EP-9 are listed below in order of priority, each with its core requirement and its pass criterion. Layers 1 to 4 and layer 8 are mandatory. Layers 5 to 7 are high priority, and layer 9 is medium priority. Each layer has its own specification page, linked in the table.

The 6-Block Entity Page Template

The 6-Block Entity Page Template is the layout every owned entity page follows, so that a person can decide and a machine can extract from the same page. The blocks run from the definition, through scope and criteria, to mistakes, proof and a next step on the decision ladder.

The 15-point technical audit checklist

The 15-point technical audit checklist turns the nine layers into checks with a named tool or method for each. Every checkpoint has an identifier from CHK-EP-01 to CHK-EP-15, so an audit can report which item failed rather than a general impression. Each layer page repeats its own checkpoints.

The official schema template for this page

The official schema template in the specification is written for this URL. It declares a WebPage about the entity, a DefinedTerm with the canonical definition, and a DefinedTermSet for the framework, all connected by @id references in one @graph. This page publishes the same three nodes, plus a term for each of the nine layers.

{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "WebPage",
      "@id": "https://danhnolan.com/en/ep9/#webpage",
      "url": "https://danhnolan.com/en/ep9/",
      "name": "Entity Architecture and Governance Framework (EP-9)",
      "isPartOf": {
        "@id": "https://danhnolan.com/#website"
      },
      "about": {
        "@id": "https://danhnolan.com/en/ep9/#entity"
      }
    },
    {
      "@type": "DefinedTerm",
      "@id": "https://danhnolan.com/en/ep9/#entity",
      "name": "Entity (EP-9)",
      "description": "An entity with a clearly defined identity, a context that separates it from namesakes, links to related entities, and enough evidence for search engines, AI systems and people to understand what it is and why it can be trusted.",
      "inDefinedTermSet": {
        "@id": "https://danhnolan.com/en/ep9/#ep9"
      }
    },
    {
      "@type": "DefinedTermSet",
      "@id": "https://danhnolan.com/en/ep9/#ep9",
      "name": "Entity Architecture and Governance Framework (EP-9)",
      "sameAs": "https://doi.org/10.5281/zenodo.22934007"
    }
  ]
}

The 30-day implementation roadmap

The 30-day implementation roadmap splits an EP-9 rollout into four sprints: inventory, entity pages, schema and links, then DOM QA and verification. The order follows the Trust Chain, so no sprint depends on work that a later sprint has not yet done.

1. Sprint 1 (Days 1-7) - Entity Audit & Inventory. Identify core domain entities. Map into 4 functional groups (Offer, Problem, Method, Trust). Audit existing URLs and external brand profiles.
2. Sprint 2 (Days 8-15) - Entity Page Architecture & Content. Deploy dedicated Entity Pages using the 6-block template. Draft Answer-First definitions and disambiguation context.
3. Sprint 3 (Days 16-23) - Schema Implementation & Internal Link Flow. Inject @graph DefinedTerm JSON-LD with DOI sameAs linkage. Build 3-5 contextual internal links per entity.
4. Sprint 4 (Days 24-30) - DOM QA & AI Extraction Verification. Validate 100% DOM text parity. Perform GSC URL inspection. Monitor Google AI Overviews and ChatGPT citations.

The 100-point Definition of Done

The Definition of Done is the rubric that decides whether an entity page is ready for production. It has five evaluation domains of 20 points each: identity on its own, then the remaining eight layers in pairs, so every layer of the framework is scored somewhere in the rubric.

How EP-9 relates to DLN and SEO Master

EP-9 relates to the other published frameworks on this site by providing the entity layer they rely on. The four entity groups map to the rungs of Decision Ladder Navigation, and the sixth block of every entity page is a DLN next-step link.

In the SEO Master framework, entity disambiguation is one of the criteria of the AI-Ready pillar. EP-9 is the detailed specification of how that entity is built, linked, declared and kept consistent.

Frequently asked questions about EP-9

These are the questions readers ask most often about the Entity Architecture and Governance Framework (EP-9). Each one is answered directly from the deposited specification, in a few sentences, so that the answer can be read, quoted and cited on its own without the rest of this page.

Is EP-9 related to the NIO EP9 electric car?

No. EP-9 on this site is the Entity Architecture and Governance Framework, a specification for building entities that search engines and AI systems can identify. It has no connection to the NIO EP9 car or to any product model number. The hyphen and the full name are used to keep the two apart.

Which EP-9 layers are mandatory?

Layers 1 to 4 are mandatory in EP-9, and layer 8 is a mandatory quality check. Layers 5, 6 and 7 are high priority, and layer 9, off-site consistency, is medium priority. A page that fails a mandatory layer is not finished, whatever it scores elsewhere.

How long does it take to apply EP-9?

Applying EP-9 takes 30 days in the published roadmap, split into four sprints. Days 1 to 7 are the entity audit, days 8 to 15 build the entity pages, days 16 to 23 add schema and internal links, and days 24 to 30 verify the rendered page and watch for citations in AI answers.

Rights, reuse and how to cite this framework

Rights and reuse for this framework follow the Creative Commons Attribution 4.0 International license stated in the deposited specification. You may copy, share, adapt and build on the material for any purpose, including commercial use, as long as you give credit, link to the license and state whether you made changes.

Cite this framework as: Nolan, D. (2026). Entity Architecture and Governance Framework (EP-9): A 9-Layer Entity Specification, 6-Block Content Template, DefinedTerm Schema, and Trust Chain Standard for AI Search Extraction. Technical Working Paper, danhnolan.com Research Group. Zenodo. 10.5281/zenodo.22934007. ORCID: 0009-0007-7906-5091. Licensing terms: rights and reuse.

The deposit on Zenodo is the authoritative version. If this page and the deposited document differ, follow the document under the DOI.

References

The references below are the standards and official documentation that the layers of EP-9 apply. They are listed so that each requirement can be checked against its source rather than taken on trust, which is the same principle the seventh layer asks of every entity.

• W3C. JSON-LD 1.1, W3C Recommendation, 2020.
• W3C. RDF 1.1 Concepts and Abstract Syntax, W3C Recommendation, 2014.
• Schema.org. DefinedTerm, DefinedTermSet and sameAs.
• Google Search Central. General structured data guidelines.
• Google Search Central. Link best practices.
• IETF. RFC 6596, The Canonical Link Relation.