Jason Burns / jasonburns.co.uk
Available - taking new work Contact
Tool AI search citation readiness

AI Schema Validator
score your schema for ChatGPT, Perplexity and AI Mode.

Paste any URL. The tool fetches the page, extracts every JSON-LD block, and scores it on what AI engines actually use to cite sources: entity graph, Wikidata, named author, freshness. Not just Google rich-result eligibility.

// AI-specific schema validator

A normal schema validator tells you if the code is valid. This checks whether it is useful.

Schema.org and Google's Rich Results Test are still worth using. They tell you whether the markup is valid and whether Google can show a visible search enhancement.

This tool answers the next question: can an AI search engine connect the page to a real entity, named author, publisher, sameAs graph and extractable answer blocks? That is the gap generic validators miss.

It checks

JSON-LD blocks, @id references, Person and Organization links, FAQ depth, dates and sameAs quality.

It does not promise

AI Overview or ChatGPT citation. Schema helps clarity. The page still needs crawl access, proof and useful visible copy.

Server-side fetch. No data is logged or stored.

Schema scored for citation in

Scoring weights reflect what each engine actually rewards - Wikidata, named-author E-E-A-T, @graph linkage, freshness. Not just Google rich-result eligibility.

  • ChatGPT by OpenAI ChatGPT
  • Claude by Anthropic Claude
  • Perplexity by Perplexity AI Perplexity
  • Gemini by Google Gemini
  • Copilot by Microsoft Copilot
  • Llama by Meta Llama
  • Grok by xAI Grok
  • Mistral by Mistral AI Mistral
  • DeepSeek by DeepSeek DeepSeek
// why AI schema is different

Scored for AI engines, not just Google rich results.

Generic schema validators check syntax against schema.org and stop there. AI search engines (ChatGPT, Perplexity, Claude, Gemini, Google AI Mode) need a deeper set of signals before they will cite a source - the ones below.

Wikidata identifier check

Wikidata Q-numbers are the strongest entity-disambiguation signal AI engines use. Without one, two same-named businesses bleed into each other in AI answers. This tool flags Wikidata absence as a Tier-1 issue, not a nice-to-have.

sameAs link tiering

Not all sameAs links are equal. Tier-1 (Wikidata, gov.uk, Companies House) > Tier-2 (LinkedIn, Crunchbase, G2) > Tier-3 (social only). The tool counts each tier and warns when sameAs is social-only - the most common entity-graph weakness.

Entity graph integrity

AI engines traverse schema via @id references. This tool walks every @id reference, errors on broken refs, errors on duplicate @ids, and warns when 3+ blocks ship with zero cross-references (siloed nodes that look like unrelated entities).

Author-as-Person enforcement

Plain string authors ("author: Jane Smith") are invisible to AI engines - they cannot resolve the writer's identity. The tool flags this as an error on Article/BlogPosting/NewsArticle nodes, even though generic validators pass it.

Substantive FAQ detection

FAQPage with one Q&A or single-line answers is gaming the schema for a rich-result hint without the substance. The tool counts substantive answers (>=80 chars) and requires >=4 for the citation credit.

Freshness as a first-class signal

Industry analysis (PromptAlpha, Ziptie) puts freshness around 30 to 40 percent of Perplexity ranking. The tool reads dateModified from your schema and credits content under 30 days, soft-credits under 90 days, and penalises year-old dates.

// schema questions

What this validator is really checking.

What is an AI schema validator?
An AI schema validator checks whether JSON-LD helps answer engines understand the page entity, author, publisher, sameAs links and useful question-answer blocks. It goes beyond syntax and rich-result eligibility.
Is this the same as the Schema.org validator?
No. The Schema.org validator checks whether markup follows the vocabulary. This tool checks whether the structured data is useful for AI search extraction, entity disambiguation and citation readiness.
Does schema guarantee AI Overview or ChatGPT citation?
No. Schema is a clarity layer, not a citation guarantee. A page still needs crawl access, useful visible content, a clear author and enough authority to be selected as a source.
// what only a human ships

The score tells you what is broken. A consultant ships the fix.

Free tools are good at diagnostics. They cannot get you a Wikidata Q-number, rebuild your entity graph across every template, or seed the citations that actually move AI rankings. That is the consultancy side.

GEO Consulting

Citation-first AI search programme

Entity graph rebuild, Wikidata setup, @id-linked schema across every template, citation seeding on Reddit / LinkedIn / podcasts. The full programme this score is measuring.

See the GEO service AI Consulting

When schema is one piece of a bigger AI gap

If your schema is broken because the brand has no AI strategy at all, fixing the JSON-LD is treating the symptom. AI consulting covers the wider plan - AI search, internal AI workflows, AI-led content.

See AI Consulting SEO Management

Schema fixes inside a full SEO retainer

For sites that need schema, technical SEO, content and authority work landed together rather than as separate projects. Monthly retainer with senior delivery, not account-manager triage.

See SEO Management Technical SEO

When the schema problem is template-level

If schema is missing, duplicated or disconnected across templates, the fix belongs in technical SEO. I map the issue and turn it into developer-ready tickets.

See Technical SEO
// next step

Want this run across every template on your site?

The tool checks one URL at a time. A real AI schema audit covers every template, builds the entity graph with proper @id linkage, gets you a Wikidata Q-number and ships the fixes. 30-min call to scope it, no obligation.

Book a 30-min call See the full GEO service