Is your structured data an entity graph — or just scattered facts?
Every free tool checks whether your schema is valid. None of them check whether it works. Schema X-Ray reads the hidden, machine-readable facts on your site and grades the entity graph AI actually relies on — how your Organization, logo, profiles and products connect by @id. Live, from your real page. Instant letter grade, full breakdown, fixes in order.
We grade the graph, not the grammar
A validator confirms your schema is spelled correctly. This reads whether the facts join up into one picture of your business — measured live, the moment you run it.
We read every structured-data block
We fetch your homepage source and pull out every JSON-LD block — the hidden, machine-readable facts about your business. We handle arrays and @graph sets, and count any block that fails to parse.
We rebuild your entity graph
This is the part no free validator does. We map how your entities reference each other by @id — Organization to logo, to founder, to your social profiles — and flag the ones left as unconnected islands or broken links.
We grade what actually matters
Consolidation, cross-referencing, placement, Organization completeness, Wikidata-linked expertise, and shopping-agent readiness — each scored, with your fixes in priority order. A letter grade, not a pass/fail.
What a valid-but-thin graph quietly costs you
Valid isn’t the same as complete
Every validator tells you whether your schema parses. None tell you whether your Organization has a logo, links its real-world identity, or connects to the rest of your graph — the things that decide rich-result eligibility.
Disconnected facts don’t build trust
If your entities never reference each other by @id, AI reads them as unrelated scraps instead of one confident picture of your business. Consolidation and @id links are how the graph becomes legible.
The shopping-agent layer is arriving
AI shopping agents, Microsoft’s NLWeb and Google’s Universal Commerce Protocol all read structured data to quote and recommend products. Priced, identified, brand-linked Product markup is the price of admission.
What structured data does and doesn’t do
Plenty of tools imply that adding schema gets you cited by ChatGPT or Google’s AI. We won’t. A May-2026 Ahrefs causal study — testing sites that added structured data against those that didn’t — found no measurable uplift in AI citations from schema alone. We build this tool on what’s verified, so it grades only what structured data genuinely controls.
Three things it does control, and what each dimension of your grade is really measuring:
- Rich-result eligibility. Google needs a well-formed Organization — name, URL, logo, contact — before it will show a logo, sitelinks or a knowledge panel. That’s the Organization-completeness and consolidation scores.
- Entity disambiguation. sameAs links and Wikidata-linked knowsAbout topics tie your site to your verified real-world identity, so AI isn’t guessing which business or subject you mean. That’s the @id cross-referencing and knowsAbout scores.
- The commerce data substrate. AI shopping agents, Microsoft’s NLWeb and Google’s Universal Commerce Protocol read Product markup — priced, identified, brand-linked — to quote and recommend you. That’s the shopping-agent-readiness score.
Source: Ahrefs, “We Tracked 1,885 Pages Adding Schema. AI Citations Barely Moved.” — causal study, 11 May 2026. Every score in this tool is computed directly from your live page source; nothing is estimated.
Fair questions
How is this different from a schema validator?
A validator answers one question: does your JSON-LD parse without errors? Schema X-Ray answers the questions that actually affect results — is your data consolidated into one connected graph, do your entities reference each other by @id, is your Organization complete enough for a Google knowledge result, and is your expertise tied to real-world identifiers. It grades the entity graph, not just the syntax.
Will this change anything on my website?
No. It reads your homepage source once, the same way a search crawler would, and grades what it finds. Nothing is changed, installed or stored on your site.
Does better structured data get me cited by ChatGPT or Google’s AI?
Not directly, and we won’t pretend otherwise. A May-2026 Ahrefs causal study found no AI-citation uplift from adding schema. What a strong entity graph genuinely earns is Google rich-result and knowledge-panel eligibility, confident entity disambiguation, and a clean data feed for the AI shopping agents and commerce protocols now coming online. Those are real, and they’re what this tool measures.
What are @id links and why do they matter?
An @id is a stable handle you give an entity — say your Organization — so other entities can point back to it. When your WebSite says its publisher is that same @id, and your logo and social profiles hang off it too, AI can follow one connected graph. Without @id links, each fact stands alone and the machine has to guess how they relate.
Do I need to enter my email?
No. This check is deliberately open — run it as many times as you like. If you want the full picture across all five AI-visibility categories, that report is one click away.