AI Visibility Audit

Scoring Categories

1. 01

   Entity Stability

   10 points

   What is evaluated

   Whether the business, its founder, and its products are defined as distinct, named entities with consistent canonical names, correct Schema.org types, and stable @id values across all surfaces.

   Why it matters

   AI systems learn associations from consistently named entities. An entity that appears under multiple names or types across its own properties cannot produce a coherent representation — and cannot be reliably cited.

   What failure looks like

   • —Organization name varies across domains and pages
   • —No Person entity declared for the founder
   • —Products referenced without consistent parent-child relationships
   • —Schema @id values differ between pages for the same entity
   • —No entity registry or equivalent naming governance
2. 02

   Category Ownership

   10 points

   What is evaluated

   Whether the business has published a structured body of content that defines authority within a specific topic domain — a deliberate cluster architecture with pillar, depth, and definition layers that establishes a clear category claim.

   Why it matters

   Topic co-occurrence is one of the primary mechanisms by which AI models associate entities with categories. If the entity name co-occurs consistently with a topic domain across many documents, the model builds association between entity and category.

   What failure looks like

   • —No topic cluster architecture — only isolated articles
   • —Content does not define a specific category or domain
   • —No pillar article establishing category ownership
   • —Thin content depth — fewer than 5 articles per topic cluster
   • —No definition pages capturing the category's core terminology
3. 03

   Schema Graph

   10 points

   What is evaluated

   Whether JSON-LD schema is correctly and consistently implemented across the site — with stable @id values, locked entity types, canonical author and publisher references, and no contradictions between schema and page copy.

   Why it matters

   Structured data reduces inference burden for AI parsers. Without it, AI systems must infer entity type, relationships, and attributes from prose alone — a weaker signal. With correct schema, entity facts are machine-readable and explicitly declared.

   What failure looks like

   • —No JSON-LD schema present
   • —Schema types don't match entity types (e.g., methodology typed as Article)
   • —Author and publisher fields duplicate entity definitions instead of using @id references
   • —Placeholder dates in schema or synthetic freshness metadata
   • —@id values are not stable or resolve to non-canonical URLs
   • —Schema contradicts page copy on entity description or classification
4. 04

   Knowledge Index

   10 points

   What is evaluated

   Whether the site's internal link architecture reinforces the entity graph — specifically whether product pages, content pages, concept pages, and reference pages cross-link coherently to the hub pages and to each other.

   Why it matters

   Internal links signal how entities relate and which pages are authoritative within the domain. A well-linked internal architecture concentrates authority signals at hub pages and allows parsers to traverse the entity graph from multiple entry points.

   What failure looks like

   • —Pages exist with no inbound internal links
   • —Concept and content pages do not link back to the ecosystem hub
   • —No cross-linking between related concept and reference pages
   • —Product domains do not link back to the parent organization domain
   • —Authority pages (ecosystem, founder) are not referenced from content pages
5. 05

   Continuous Signal Surfaces

   10 points

   What is evaluated

   Whether the entity maintains a growing, consistent surface of pages and cross-domain references — including directory profiles, cross-domain product references, and sameAs schema references pointing to real identity profiles.

   Why it matters

   A single self-declaration is a weak authority signal. A continuous, expanding surface of consistent references across domains and page types compounds retrieval signal over time. AI systems build citation confidence from the breadth and consistency of references they encounter.

   What failure looks like

   • —Entity defined only on its own domain with no cross-domain reinforcement
   • —Product domains do not reference the parent organization
   • —No directory profiles with canonical entity names
   • —sameAs URLs in schema point to product domains or non-existent profiles
   • —No ongoing publication of consistent, linked content extending the surface

Score Interpretation

• 45–50
  Strong

  Entity is well-defined, structured, and cross-referenced. AI retrieval is likely to be consistent.

• 38–44
  Good

  Core structure is in place. Specific gaps in citation surfaces or topic depth are limiting citation reliability.

• 30–37
  Moderate

  Entity architecture exists but has inconsistencies. Structured data or citation surfaces need attention.

• 20–29
  Low

  Significant structural gaps. Entity is unlikely to be retrieved reliably by AI systems.

• 0–19
  Critical

  Entity is undefined or fragmented. AI systems cannot form reliable associations around this entity.

Who This Is For

• Software companies

  Products with no entity definition, no schema, and no cross-domain reinforcement — well-built tools that AI systems cannot reliably recommend.

• Research publications

  Publications with strong content but no entity architecture — AI systems cannot associate the publication name with a specific author or organization entity.

• Consultancies and studios

  Service businesses that depend on category authority — where appearing in AI answers to category questions has direct business value.

• Organizations with multiple product domains

  Entities where the parent-child relationship between organization and products is undefined — resulting in authority fragmentation across domains.

Frequently Asked Questions

What does an AI Visibility audit include?

A Jonomor AI Visibility audit is a 50-point structured evaluation across five categories: Entity Stability, Category Ownership, Schema Graph, Knowledge Index, and Continuous Signal Surfaces. It produces a scored diagnostic report identifying where AI retrieval gaps exist and which implementation actions will close them.

How long does an AI Visibility audit take?

An AI Visibility audit begins with a baseline retrieval cycle run across multiple answer engines, followed by scoring and diagnosis. The initial audit report is typically delivered within one to two weeks of engagement start.

Who is an AI Visibility audit for?

AI Visibility audits are designed for software companies, research publications, consultancies, and multi-product organizations where AI retrieval gaps are structural — entity fragmentation, absent schema, authority isolation — rather than purely content volume problems.

What is the difference between an AI Visibility audit and an SEO audit?

An SEO audit evaluates ranking signals for search engines. An AI Visibility audit evaluates how well an organization's entities are defined, typed, and structured for recognition and citation by AI answer engines like ChatGPT, Perplexity, Gemini, and Copilot.

What happens after the audit?

The audit produces a scored baseline and a prioritized implementation roadmap. Engagements can continue into full authority architecture work, or the audit can be delivered as a standalone diagnostic.