How can websites represent entities and their relationships to help search systems understand them?

ProcrustesPR

To help search systems understand entities such as people, organizations, and products, websites should use consistent names and identifying information for each entity. Content, main pages, category links, and structured data should clearly explain the relationships between them. Beyond identifying “who or what this is,” websites should explain “how it relates to other entities and in which contexts those relationships apply.” Check separately whether the relationships are accurate, the markup is valid, and search systems can access the content.

ProcrustesPRProcrustesPRCan a One-Page Website Do SEO?

List entities first, then verify identities and relationships

In a Search Engine Journal article, Harry Clarkson-Bennett recommends starting with the organization, mapping related people, products, services, locations, and topics, and then finding supporting evidence for each relationship. This entity map can help identify inconsistencies between how the organization is represented externally and how it sees itself.

  • Use consistent names and identifying information when the same person or brand appears on different pages.
  • If products use SKUs, ISBNs, or GTINs, use those identifiers consistently wherever the products appear rather than relying solely on similar descriptions to distinguish them.
  • Clarify brand ownership, product families, and service locations rather than merely listing names.

Keeping identifiers consistent is a technical recommendation from Helen Pollitt, not an official requirement that every website use all of these identifier types.

Show relationships through main pages and categories

Pollitt recommends creating comprehensive main pages for key people, brands, or products. For example, linking an article back to an author page can show the relationship between the author and their work. The author page can then link to categories covering their areas of expertise and information about their work published elsewhere, helping clarify who the author is.

Categories should also have a clear hierarchy. For example, “Clothing” might contain “T-shirts” and “Dresses,” with category-page links and breadcrumbs showing parent-child relationships. Within a page, use heading levels to organize topics and sentences or lists to explain relationships between entities, rather than simply repeating their names.

Use structured data within its intended scope

Google’s product variant documentation states that when products vary by attributes such as size or color, ProductGroup, along with variesBy, hasVariant, and productGroupID, can be used in addition to Product markup to group variants under the same parent. This approach is intended for product variants and should not be applied to all related products.

Relationships also need context. Bill Hunt’s “integrity graph” concept emphasizes recording facts along with the contexts in which they apply—for example, the markets in which a service is available, the branches that provide it, and whether information applies globally or only locally. This is an approach to organizing information proposed by the author, not a Google markup requirement.

Validate markup and access first, then check entity relationships

According to Google’s documentation, you can first validate the code with the Rich Results Test and fix critical errors. Then deploy a small number of pages and use the URL Inspection tool to check how Google renders them. Confirm that login restrictions or relevant access blocks are not preventing access.

Markup testing and rendering checks serve different purposes. When reviewing entity relationships, also return to the entity map and check names, identifiers, categories, and supporting evidence for each relationship. For ecommerce sites that also supply product feeds, Pollitt notes that feed data should match the data on the website and elsewhere.

ProcrustesPRProcrustesPRHow to Use a Marketing Funnel: Fill Page Gaps at Each Stage, Then Build a Conversion Path

Entities and relationships: perspectives from across the web

How do search systems determine whether mentions in different places refer to the same entity?

Helen Pollitt notes that search engines try to compare signals such as names, addresses, social profiles, topical context, and co-occurrence patterns to determine whether multiple mentions across the web refer to the same entity. These signals should support one another. Related names appearing in the same passage do not, on their own, establish a specific relationship between them; it is still advisable to state the relationship explicitly and verify the facts.

Does important content have to use server-side rendering?

Not necessarily. Pollitt recommends making important content easily accessible to search and AI bots and presents server-side rendering as one approach. She also notes that Googlebot can render JavaScript, but reducing reliance on JavaScript may still be good practice. The priority is to confirm that bots can access the content they need to identify, rather than treating one rendering method as a mandatory requirement for every website.

Do non-critical issues in structured data testing mean a page is ineligible for rich results?

No. According to Google’s product variant documentation, a page may still be eligible for rich results even if it has non-critical issues. Google nevertheless recommends fixing those issues to improve structured data quality. Critical errors found during testing must be fixed. This distinction applies to the product variant markup described in that documentation, and passing the test does not guarantee that rich results will appear.

ProcrustesPRWhat types of URLs are there, and what are they used for?

References

  1. 產品子類結構化資料 (ProductGroup、Product) | Google 搜尋中心  |  Documentation  |  Google for Developersdevelopers.google.com — Discusses: Do non-critical issues in structured data testing mean a page is ineligible for rich results?
  2. Entities Are The Foundation Of Modern SEOwww.searchenginejournal.com · 2026-09-02T14:31:43+00:00 — Source used in the article above.
  3. The Integrity Graph: The Missing Infrastructure Layer For AI Visibilitywww.searchenginejournal.com · 2026-06-10T13:30:06+00:00 — Source used in the article above.
  4. How Can You Implement Entity Optimization Without Relying On Schema Markup? – Ask An SEOwww.searchenginejournal.com · 2026-06-04T08:30:30+00:00 — Discusses: How do search systems determine whether mentions in different places refer to the same entity?; Does important content have to use server-side rendering?

Related articles

Can a One-Page Website Do SEO?How to Use a Marketing Funnel: Fill Page Gaps at Each Stage, Then Build a Conversion PathWhat types of URLs are there, and what are they used for?When should you use relative or absolute paths in WordPress?What are www subdomains and relative links used for?How to Create an HTML Sitemap: From Organizing Content to Publishing and UpdatingHow to troubleshoot and fix 404 errors: identify broken URLs first, then address the causeHow to improve on-site content and structure: audit pages first, then address content gaps and internal links
ProcrustesPR
ProcrustesPR