sameAs Schema: The 12 Links That Confirm Your Entity

When someone asks Google or ChatGPT about your company, how does the machine know it has the right one? There are four businesses called Northwind in your state, two founders who share your name, and a defunct startup that used your brand in 2017. The answer sits in a small JSON-LD array called sameAs, and most sites either skip it or fill it with the wrong links.

This guide covers what sameAs schema does according to the people who define it, which 12 links belong in it (ranked), and how to keep those links from contradicting each other.

Write the JSON-LD before you collect a single link

Start with the container, because the container decides what the links describe. The schema.org definition of sameAs reads: “URL of a reference Web page that unambiguously indicates the item’s identity.” The expected value is a URL, and the property sits on Thing, which means every type inherits it. In practice you attach it to one Organization (or a subtype like LocalBusiness) on your home page, and one Person on each founder or author page.

Here is a working template. Replace the example URLs with your own and delete any line you cannot fill.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "@id": "https://www.example.com/#organization",
  "name": "Northwind Analytics",
  "legalName": "Northwind Analytics, Inc.",
  "url": "https://www.example.com/",
  "logo": "https://www.example.com/logo.png",
  "foundingDate": "2019",
  "founder": {
    "@type": "Person",
    "@id": "https://www.example.com/about/jane-doe/#person",
    "name": "Jane Doe"
  },
  "sameAs": [
    "https://www.wikidata.org/wiki/Q00000000",
    "https://en.wikipedia.org/wiki/Northwind_Analytics",
    "https://find-and-update.company-information.service.gov.uk/company/00000000",
    "https://www.linkedin.com/company/northwind-analytics/",
    "https://www.crunchbase.com/organization/northwind-analytics",
    "https://maps.google.com/?cid=0000000000000000000",
    "https://www.g2.com/products/northwind-analytics/reviews",
    "https://www.youtube.com/@northwindanalytics",
    "https://x.com/northwind",
    "https://www.facebook.com/northwindanalytics",
    "https://www.instagram.com/northwindanalytics/",
    "https://github.com/northwind-analytics"
  ]
}
</script>

Three details matter more than the list itself. The @id gives your entity a stable identifier on your own domain, so a Person object on another page can point back to it with a single line. The name must match the name on every profile you list. And each URL should be the canonical profile address, not a share link with tracking parameters.

Google’s Organization structured data guide (last updated September 2026) recommends placing this markup on your home page or a single page that describes your organization, and says you do not need it on every page. It also tells you to validate with the Rich Results Test. Run the Schema Markup Validator too, since it checks the schema.org vocabulary itself.

Do the same for people. A founder or author page should carry its own Person node with an @id, the person’s name, jobTitle, a worksFor reference pointing at your Organization @id, and a sameAs array of that person’s profiles: LinkedIn, X, a personal site, a Wikidata item if one exists, and a Muck Rack or ORCID page where relevant. Keep the person’s profiles out of the company array and the company’s profiles out of the person array. Mixing them tells a parser that Jane Doe and Northwind Analytics are the same thing, which is the opposite of what you want.

What does Google say sameAs schema does?

Keep two sets of claims apart: what Google documents and what SEO practitioners believe.

Google’s documentation is short. The Organization guide lists sameAs as a recommended property, describes it as “the URL of a page on another website with additional information about your organization,” and gives its use as a URL to your organization’s profile on “a social media or review site.” The guide says organization markup can help Google “disambiguate your organization in search results” and that some properties can influence visual elements such as the knowledge panel. It also says there are no required properties. Google’s ProfilePage documentation recommends sameAs on the Person or Organization behind a creator profile, defined as “the URL to other external profiles or home pages for the profile.”

That is the whole official story. Google does not call sameAs a ranking factor, and its AI features page says that “there’s also no special schema.org structured data that you need to add” to appear in AI Overviews or AI Mode. It does list keeping structured data consistent with visible page text as a good practice.

Microsoft has said a bit more. According to Search Engine Land, Bing’s Fabrice Canel confirmed on stage at SMX Munich that Microsoft uses schema markup to help its LLMs understand content. Nobody at OpenAI, Anthropic, or Perplexity has published comparable documentation about sameAs.

So the practitioner case for sameAs schema in AI search is an inference, and you should say so when you pitch it to your team. The reasoning goes like this: AI answers draw on search indexes and knowledge graphs, those systems resolve entities, and sameAs gives them an explicit, machine-readable map of which profiles belong to you. It is cheap insurance with a documented benefit in Google’s own disambiguation. Just do not promise a citation lift you cannot measure.

Audit every profile page before it goes in the array

A sameAs link says “this page is the same entity as me.” If the LinkedIn page lists a different founding year, the Crunchbase profile shows an old logo, and the Facebook page still carries the pre-rebrand name, you have handed the machine three contradictions and asked it to trust you.

Open every candidate profile in a tab and check six fields against your home page: exact legal and trading name, logo, one-line description, website URL (pointing to your canonical domain, https, with or without www to match), headquarters city, and founding date. For a Person, swap in full name, headshot, current title, and employer. Build a simple sheet with one row per platform and one column per field, and do not add a URL to the array until its row matches.

Then check the reverse link. Each profile should point back to your website in its own website field. A bidirectional link (your site claims the profile, the profile claims your site) is the strongest signal you can give, and it is the same logic Google uses when you claim a knowledge panel. Google’s knowledge panel verification help page asks you to sign in to one of the official sites or profiles it has associated with the entity, naming YouTube, Search Console, Twitter, and Facebook.

Assign one owner for this sheet. In most companies, social handles belong to marketing, Crunchbase to the founder, Google Business Profile to operations, and the website to engineering, which is how four versions of the same company end up online. One person should hold edit access, or at least a log of who does, for every URL in the array.

Kill the orphans while you are there. Old Twitter handles, a duplicate LinkedIn page an intern created, a Yelp listing for a closed office: merge them, close them, or at least make sure none of them appear in your sameAs array.

The ARC Tiers: how to rank the 12 links

Not every profile carries equal weight. I rank sameAs candidates with a three-tier model called ARC: Authority sources that third parties maintain, Registries that verify you exist as a business, and Channels where you publish. The order reflects how hard each link is for an impostor to fake and how often knowledge systems reuse the data.

Fill from the top when you can, but never add a link that does not exist yet just to complete the set. Seven accurate links beat twelve shaky ones.

Tier A: the three authority anchors

1. Wikidata. Wikidata launched in 2012, publishes its data under a CC0 public domain dedication, and absorbed data from Freebase after Google announced in December 2014 that it would shut Freebase down “in favor of Wikidata,” according to Wikipedia’s article on the project. Freebase fed Google’s early Knowledge Graph, which is why SEOs treat a Wikidata Q-ID as the most valuable link in the array. Google has not published how much it uses Wikidata today, so hold that as informed belief. Wikidata items must meet a notability standard: a page on a Wikimedia site, a clearly identifiable entity described by serious public references, or a structural need. Link a company that has press coverage or registry records, and expect deletion if the only source is your own site.

Once the item exists, close the loop from the Wikidata side. Add your domain under official website (P856), then add the matching profile identifiers: LinkedIn company ID (P4264), YouTube channel ID (P2397), X username (P2002), Facebook username (P2013), and Instagram username (P2003). Now your site points to Wikidata and Wikidata points back to your site and to the same profiles in your array. That symmetry is the whole point of sameAs schema, and Wikidata is the one place you can express it in a public, machine-readable graph.

2. Wikipedia. If an article exists, link it. If one does not, do not write it yourself. Wikipedia’s conflict-of-interest guideline discourages editing articles about yourself or your employer, and a deleted article leaves a worse trail than none.

3. Your government business registry. Companies House in the UK, a state Secretary of State record in the US, SEC EDGAR for public companies. These pages prove legal existence, and practitioners rarely list them, which is a quiet advantage. OpenCorporates aggregates many registries if your local one lacks stable URLs.

Tier R: registries that verify the business

4. LinkedIn company page. For B2B brands and founders, LinkedIn is often the first profile a buyer or journalist checks. Match the company name and website field exactly, and link each founder’s personal profile from their Person markup.

5. Crunchbase. Reporters and investors look up funding and founding facts on Crunchbase, so an error there travels. Claim the profile and correct the founding date and headquarters before linking it.

6. Google Business Profile. For any business with a storefront or service area, this is Google’s own verified record of you. Practitioners prefer the Maps URL with the CID parameter because it points to the listing ID rather than a name that might change. Google’s knowledge panel help points local businesses to Business Profile rather than the panel claim flow.

7. A review platform profile. Google’s Organization guide names review sites as valid sameAs targets. Pick the one your buyers trust: G2 or Capterra for software, Trustpilot for consumer brands, Clutch for agencies, Healthgrades for clinics. One strong profile beats five empty ones.

Tier C: channels you publish on

Channels rank last because anyone can open a profile under any name. They still earn their place: they show the entity is active, they often carry verification badges, and four of them double as Google’s knowledge panel verification routes.

8. YouTube. Google owns it, Google names it in the knowledge panel verification flow, and video descriptions give you room for consistent entity language. Use the @handle URL.

9. X. Also in Google’s verification list (as Twitter). Use the x.com address and make sure the bio links your domain.

10. Facebook page. Use the page URL, not a personal profile. Pages for local businesses should mirror the Google Business Profile address character for character.

11. Instagram. Include it if you post there with some regularity. A dormant account with 40 followers is a poor identity signal.

12. GitHub or your vertical’s channel. Developer tools should link the GitHub organization. Swap in what fits your field: Spotify or Apple Podcasts for a show, Behance for a design studio, TikTok for a consumer brand that lives there. For a Person, this slot often goes to an ORCID iD for researchers, an IMDb page for film people, or a Muck Rack profile for journalists.

Which sameAs mistakes work against you?

The common failure is listing pages that are not about you. A Wikipedia article about your industry, a news story that mentions you, or a parent company’s profile all fail the schema.org test of a page that “unambiguously indicates the item’s identity.” Use the subjectOf or mentions properties for press coverage instead.

The second is drift. Teams rebrand, change handles, or merge LinkedIn pages, and nobody updates the JSON-LD. Put a quarterly check on the calendar: load each sameAs URL, confirm it resolves without a redirect chain, and rerun the consistency sheet.

The third is markup that disagrees with the visible page. Google’s AI features guidance lists matching structured data to visible text as good practice. If your schema says “Northwind Analytics, Inc.” and your footer says “Northwind Data Co.,” fix the footer or the schema so they agree.

The fourth is scattering Organization markup with different @id values across dozens of pages, which can read as several entities. Keep one canonical Organization node and reference it.

If you want an outside view of how answer engines see your brand today, the free AEO rating from Instant Press checks entity signals like these alongside citations.

Your next step: open a sheet, list the 12 slots from the ARC table, paste the URL you have for each, and fill the six consistency columns before you touch a line of code.