Building a Connected Schema Graph
Most websites implement structured data as isolated blocks — one Organization schema on the homepage, one Article schema on blog posts, maybe a LocalBusiness schema on the contact page. These disconnected blocks miss the entire point of structured data.
The Graph Concept
Google's structured data documentation and related patents (including US Patent 8,868,544) describe structured data not as individual declarations but as a graph — a network of entities connected by typed relationships. The @id property is the key mechanism that connects this graph.
The @id Architecture
Every entity on your site should have a unique @id — a URI that serves as its permanent identifier within your schema graph.
Pattern: https://yourdomain.com/#organization, https://yourdomain.com/#website, https://yourdomain.com/page/#webpage
Once entities have @ids, they reference each other:
The Core Graph Structure
- Organization (root entity) — @id:
/#organization
- Has url, name, logo, sameAs, contactPoint
- WebSite — @id:
/#website
- publisher → references Organization @id
- Has url, name, potentialAction (SearchAction)
- WebPage (per page) — @id:
/page/#webpage
- isPartOf → references WebSite @id
- Has url, name, description, breadcrumb
- Article (per content page) — @id:
/page/#article
- mainEntityOfPage → references WebPage @id
- publisher → references Organization @id
- author → references Person @id
Supporting Entities
- Person (author) — @id:
/#person-firstname-lastname
- worksFor → references Organization @id
- Has name, jobTitle, sameAs, knowsAbout
- BreadcrumbList (per page) — @id:
/page/#breadcrumb
- itemListElement → ordered list of page hierarchy
Common Graph Errors We Fix
- 1Orphaned entities — Schema blocks with no @id references to other entities. They float in isolation.
- 2Broken @id references — Entity A references Entity B's @id, but Entity B does not exist in the graph.
- 3Inconsistent @id patterns — The same entity uses different @ids on different pages, creating duplicates.
- 4Missing isPartOf chains — WebPages that do not connect to the WebSite entity.
- 5Publisher/author disconnections — Articles without proper references to the publishing Organization and authoring Person.
How We Implement
Our Structured Data Integrity dimension scores your schema graph across:
- Connectivity — What percentage of entities reference at least one other entity?
- Completeness — What percentage of recommended properties are filled?
- Consistency — Are @ids used uniformly across all pages?
- Validity — Zero errors in structured data testing tools?
We deploy the complete schema graph via our implementation snippet. Every page receives the correct entities with full @id connectivity, and we validate the graph across your entire site.
The result: search systems understand not just what each page is about, but how every entity on your site relates to every other. That is the foundation for Knowledge Panel eligibility, rich results, and enhanced SERP features.