Glossary

SameAs schema tells machines every profile belongs to the same business

SameAs is a schema.org property that lists the other pages on the web describing the exact same real-world entity as the one being marked up: a Wikipedia article, a Wikidata item, a LinkedIn company page, an official social profile. Placed inside a business's structured data, it tells search engines and AI systems that scattered mentions across the web all refer to one entity, not several similarly named ones.

Definition

sameAs is a schema.org property, most often added to an Organization, Person, or LocalBusiness entry as an array of URLs, each pointing to another page that represents the identical entity from a different source. It is a disambiguation signal: it does not add new facts about a business, it declares that a set of separately-discovered pages are all describing the same one.

Why a single URL is not enough

A business's own website already states its name, category, and location in its own Organization markup, using the url property to point back at itself. That is one page's word for it. A system trying to verify who a business actually is has to decide whether to trust a single self-reported source, or whether independent sources agree.

sameAs supplies the independent sources. It lists the business's presence elsewhere on the web, places it did not build itself and cannot unilaterally edit, so an outside system has something to cross-check the self-reported claim against.

What actually belongs in the array

The strongest entries are the ones an outside system can independently verify. A Wikidata item, identified by its own Q-number, is considered the single strongest sameAs target, because Wikidata feeds directly into the knowledge graphs that both search engines and AI systems consult. A Wikipedia article, where one exists and meets notability standards, is similarly strong.

Below that: verified official social profiles, a Crunchbase or industry-directory listing, and any other page a business genuinely controls and keeps current. What does not belong is an abandoned profile, an unclaimed listing with the wrong details, or a page the business no longer maintains. A stale or contradictory sameAs entry is worse than a shorter, accurate array, because it hands a verifying system a source that disagrees with the rest.

One entity, cross-checked

sameAs is the line of markup that turns a set of separate mentions into one entity a machine is willing to name with confidence.

sameAs versus url

The two properties point in opposite directions. url is a page's own canonical address, the site speaking for itself. sameAs points outward, to other pages, on other domains, that describe the same thing from the outside. A homepage carries its own url; its sameAs array lists everywhere else the business is verifiably represented.

Confusing the two is a common markup mistake: listing internal pages of the same site under sameAs adds nothing, because a site cannot independently corroborate itself. The value of the property comes entirely from pointing somewhere else.

0.664

Correlation between branded web mentions and being named by AI, against 0.218 for backlinks across 75,000 brands. The authors caveat that correlation is not causation, and the sample skews to established domains.

Ahrefs, 75,000 brands

What this changes for AI visibility specifically

A ranking algorithm can place a page reasonably well from partial signals. An AI system deciding whether to name a specific business in an answer has a narrower decision to make: cite this entity with confidence, or leave it out in favour of one it can verify more easily. sameAs is one of the cheapest ways to make a business the easier one to verify.

It is not, on its own, a ranking or citation guarantee. It is a precondition: it gives a system that already wants to check a claim the means to check it quickly, against sources the business does not directly control.

Common questions

What is the sameAs property?

A schema.org property that lists other pages on the web describing the exact same real-world entity as the one being marked up. It takes an array of URLs: a Wikipedia article, a Wikidata item, a LinkedIn company page, official social profiles, anywhere else the same business is represented.

Which pages should go in a business's sameAs array?

Official, verified, and actively maintained profiles only: a Wikidata item if one exists, a Wikipedia article if the business is notable enough to have one, the business's own verified social profiles, and listings on directories like Crunchbase or industry-specific registries. Abandoned or unclaimed profiles do more harm than leaving the array shorter.

What is the difference between sameAs and url?

url is the entity's own canonical page, the site describing itself. sameAs points outward to other pages, on other domains, that describe the same entity from the outside. A business marks up its own homepage with url and its Wikidata, LinkedIn, and Wikipedia presence with sameAs.

Does adding sameAs improve Google rankings directly?

No. It is not a ranking factor. What it does is remove ambiguity: it tells a system building an entity record that several separately-crawled pages describe one thing, not several similarly named things. That disambiguation is a precondition for confident citation, not a ranking signal in itself.

Why does sameAs matter more for AI systems than it used to for search?

An AI system answering a question about a business has to decide, in the moment, whether to name it with confidence. A business it can cross-verify across multiple linked, consistent sources is a safer thing to cite than one it can only find described once, in one place, with no way to confirm the description elsewhere.

Related
Next

See whether your site's structured data actually points anywhere verifiable.

Get a visibility auditor talk to the studio