The sameAs Signal Chain

Explicit vs. Implicit Reconciliation

In Session 7.1, we described how Google reconciles entities by matching signals across sources. Most of that matching is implicit: Google crawls your website, crawls your LinkedIn page, notices the same company name and URL, and infers they are the same entity.

Implicit reconciliation works, but it is fragile. Name variations, URL differences, or missing attributes can break the inference. The sameAs property in schema.org gives you a way to make reconciliation explicit. Instead of hoping Google figures it out, you tell Google directly: "This entity on my website is the same as this entity on LinkedIn, this entity on Facebook, and this entity on Wikidata."

How sameAs Works

In your Organization (or Person) JSON-LD markup, the sameAs property takes an array of URLs. Each URL points to a profile or page on another platform that represents the same entity.

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Company Name",
  "url": "https://yourcompany.com",
  "sameAs": [
    "https://www.linkedin.com/company/yourcompany",
    "https://www.facebook.com/yourcompany",
    "https://www.instagram.com/yourcompany",
    "https://twitter.com/yourcompany",
    "https://www.wikidata.org/wiki/Q12345678",
    "https://www.crunchbase.com/organization/yourcompany"
  ]
}

Google follows these URLs and checks whether the information on those pages corroborates your entity claims. If your schema says your name is "PT Arsindo Perkasa" and your LinkedIn company page says the same, that is a confirmed reconciliation signal.

The Signal Chain Model

Think of sameAs as building a chain. Your website is the anchor. Each sameAs link adds a link in the chain. The chain is only as strong as its weakest link.

The solid arrows represent your sameAs declarations (outbound from your schema). The dashed arrows represent backlinks from those profiles to your website. Both directions matter. A sameAs link to a LinkedIn page that does not link back to your website is weaker than one that does.

Which Profiles to Include

Not every URL belongs in your sameAs array. Include only profiles that represent the entity itself, not pages that merely mention it.

Note that Google Business Profile URLs should not be in sameAs. Google already knows about your GBP. Adding it to sameAs is redundant and can confuse the reconciliation engine.

Bidirectional Verification

The strongest reconciliation happens when both sides confirm the relationship. Your website's sameAs points to LinkedIn. Your LinkedIn page's website field points back to your website. This bidirectional verification is what Jason Barnard of Kalicube calls "corroborative claims." Google sees both sides asserting the same relationship and gains high confidence.

When only one side makes the claim, Google still registers it, but with lower confidence. If your schema includes a sameAs link to a Facebook page that has a completely different business name and no link back to your site, that signal is noise rather than signal.

Wikidata: The Highest-Value sameAs Target

Among all sameAs targets, Wikidata carries the most weight. Wikidata is an open, structured knowledge base maintained by the Wikimedia Foundation. It feeds directly into Google's Knowledge Graph. A Wikidata item with property P856 (official website URL) pointing to your domain, combined with a sameAs link from your schema to that Wikidata item, creates a strong bidirectional reconciliation loop through an authoritative intermediary.

If your entity qualifies for a Wikidata entry (covered in Module 4, Session 4.7), making it part of your sameAs chain should be a priority.

Further Reading

• sameAs property for Knowledge Graph entities - Schema.org GitHub discussion on sameAs implementation patterns
• Using @id in Schema.org Markup for SEO, LLMs, and Knowledge Graphs - Momentic on connecting schema entities
• Google Knowledge Graph: Definition and How to Use It - Kalicube's comprehensive Knowledge Graph guide
• Measurable Impact of Scaling Entity Linking - Schema App on entity linking results