Organization schema: the correct block, and how AI search reads it

Organization schema is a block of JSON-LD on your site that states who your business is in machine readable form: the name, the legal name, the website, the logo, the address, how to contact you, and the other profiles that belong to the same company. Put one block on your site, keep it identical everywhere it appears, and make every value match what a human can read on the page. It helps engines resolve your pages to one entity and get your basic facts right. It does not make ChatGPT, Gemini, Perplexity, Claude or Google AI Overviews recommend you, and no schema property exists that would.

Last verified: 08 October 2026

What Organization schema is, in one paragraph

Schema.org defines a vocabulary of types and properties for describing things on the web. Organization is the type for a company, a charity, a school or any other body of people. You write it as JSON-LD inside a script tag in your HTML. Google's structured data documentation names JSON-LD as its recommended format, which is why nobody should be hand writing microdata attributes in 2026. The point of the block is disambiguation: it tells a machine that this website, this logo, this address and these social profiles all belong to one named thing.

A copyable example

Change the values, keep the shape. This validates as JSON and uses only properties that exist in the schema.org vocabulary.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "@id": "https://example.com/#organization",
  "name": "Example Analytics",
  "legalName": "Example Analytics Private Limited",
  "alternateName": "Example",
  "url": "https://example.com/",
  "logo": {
    "@type": "ImageObject",
    "url": "https://example.com/logo.png",
    "width": 512,
    "height": 512
  },
  "description": "Example Analytics builds reporting software for mid sized retailers.",
  "foundingDate": "2019-04-11",
  "email": "hello@example.com",
  "telephone": "+91-80-4000-0000",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "4th Floor, 21 Church Street",
    "addressLocality": "Bengaluru",
    "addressRegion": "KA",
    "postalCode": "560001",
    "addressCountry": "IN"
  },
  "contactPoint": [
    {
      "@type": "ContactPoint",
      "contactType": "customer support",
      "email": "support@example.com",
      "telephone": "+91-80-4000-0001",
      "availableLanguage": ["en", "hi"]
    }
  ],
  "sameAs": [
    "https://www.linkedin.com/company/example-analytics",
    "https://github.com/example-analytics",
    "https://www.crunchbase.com/organization/example-analytics"
  ]
}
</script>

Paste it into the head of your site template, or into the body if your platform makes that easier. Both are valid.

Field by field: what to put, and what to leave out

Property	Put in	Common mistake
name	The name customers use. One name, not a tagline.	Stuffing the name with keywords, which reads as spam and conflicts with your own pages.
legalName	The registered entity name, exactly as filed.	Leaving it out when it differs from the trading name, which is the single most useful thing this block can resolve.
url	Your canonical home page, with the same protocol and the same www or non www form you actually serve.	Mixing http and https, or www and non www, between schema and canonical tags.
logo	A square image on a URL that returns 200 to an anonymous request.	A logo behind a login, hotlinked from a CDN that blocks bots, or a transparent PNG that renders as nothing.
@id	A stable fragment URL such as https://example.com/#organization, reused by your other schema blocks.	Omitting it, so each page declares a fresh unconnected organization.
sameAs	Profiles you control or that are genuinely about you: LinkedIn, GitHub, Crunchbase, a Wikipedia or Wikidata entry if one exists.	Padding the list with pages that are not about your company. A wrong entry here actively confuses the matching you are trying to help.
address, contactPoint	Values identical to the ones printed on your contact page.	An old address in schema and a new one in the footer, which makes both less trustworthy.
foundingDate, founder, numberOfEmployees	Only if true and only if you will maintain them.	A headcount from three years ago. A stale fact is worse than a missing one.
aggregateRating on your own Organization	Nothing. Leave it out.	Self declared ratings. Google's review snippet guidance treats self serving reviews about an organization as ineligible, and this is a reliable way to get a manual action rather than a star.

Where to put it, and how many times

One Organization block describing the whole company, served sitewide from your template. Do not write a different organization on every page. If you need other types, link them to the same @id instead of repeating the organization inline:

• WebSite with publisher pointing at https://example.com/#organization.
• Article or BlogPosting with publisher pointing at the same id.
• Product with brand or manufacturer pointing at the same id.
• A physical shop or office gets its own LocalBusiness block on its own location page, with parentOrganization pointing at the same id.

Two rules decide whether any of this works. Every value must also be visible to a human on the page, and the markup must be in the HTML the server returns, not injected by JavaScript that a crawler may not execute. If your tag manager writes the block after load, assume some engines never see it.

What it does for AI search

Structured data is how you hand an assistant your facts in a form it can read without guessing. What it does is remove reasons for a machine to get you wrong:

• Disambiguation. If another company shares your trading name, a legal name plus a matching address plus consistent sameAs profiles gives an engine something to separate you by.
• Fact accuracy. When an assistant states your location, your founding year or your support email, a clean block is one more consistent source agreeing with your page text.
• Cheap consistency. It is a single file edit, it is verifiable in minutes, and it stops the drift where your footer, your contact page and your schema each say something different.

Here is what it is for, and what the rest of the work is. No assistant has published that Organization markup affects whether a business is recommended, and none of the five we track documents it as a ranking or selection input. Assistants answer a recommendation question by reading pages and sources, and the sources that move those answers are usually not yours: directories, trade press, review sites, public documentation. Schema cannot substitute for being written about somewhere you do not own. If a page has nothing a reader would quote, correct markup will not rescue it. Our page on how structured data affects whether AI quotes you goes further into that distinction, and technical GEO setup covers the rest of the plumbing.

How to check it worked

1. Validate the syntax. Paste the block into the Schema Markup Validator at validator.schema.org. It checks your markup against the schema.org vocabulary rather than against one search engine's features.
2. Check Google's view. Run the page through Google's Rich Results Test. Remember it reports only on features Google supports, so "no rich results" is not the same as "invalid".
3. Read the served HTML. Fetch the page with a plain request, not a logged in browser, and confirm the ld+json block is in the response body.
4. Open the logo URL in a private window. If it needs a session, bots cannot see it.
5. Watch Search Console. The unparsable structured data report will tell you about syntax errors it finds across the site, which catches template breakages you would otherwise miss.

Asked and answered: organization schema

Is Organization schema required?

No. Nothing breaks without it and no engine demands it. It is a low cost, low risk way to state your identity consistently, which makes it worth doing on almost any site, but it is not a requirement and it is not a ranking factor anyone has documented.

Where exactly should the Organization block live?

Once, sitewide, from your template, describing the whole company. Give it a stable @id and reference that id from your other blocks instead of repeating the organization. If you only want it in one place, the home page is the conventional choice, but sitewide is simpler to maintain.

What is the difference between Organization and LocalBusiness schema?

LocalBusiness is a subtype of Organization for a business customers visit, so it adds things like opening hours and geo coordinates. Use Organization for the company as a whole and a LocalBusiness block on each location page, joined with parentOrganization. If you have no premises customers visit, you do not need LocalBusiness at all.

Does sameAs help an assistant identify my business?

It can help, because it points at other pages about the same entity, and that is exactly the matching problem a machine has. But no engine publishes how much weight it carries, so treat it as a hint rather than a mechanism. List only profiles that are genuinely yours. A wrong entry does the opposite of what you intended.

Will adding Organization schema get us named in ChatGPT or AI Overviews?

No, and anyone promising that is guessing. Markup helps a machine read your identity correctly. Being named in a recommendation answer depends on what the engines read about you, most of it on sites you do not control. Add the schema because it is cheap and correct, then measure whether the engines name you and work on the reasons they do not. Our guide on how to measure AI visibility sets out that method.

Can I use multiple Organization blocks for multiple brands?

Yes, if they are genuinely separate entities, each on its own site or its own section with its own @id. Link them with parentOrganization and subOrganization. What you should not do is declare two unrelated organizations on the same page, which leaves a machine to guess which one the page is about.

Where AI Knows Us fits

Schema is a one time fix. The question it cannot answer is whether an assistant names you when a buyer asks. AI Knows Us measures that across ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews, asks every question without naming the business, finds the reasons it is not named, and then fixes them by writing pages and getting the business onto the sites those engines read. A free scan needs no account. Paid plans are not open yet and the product takes no payment today. If all you came for was the schema block, copy it from above and close the tab. You do not need a product for this.

What to do first

Copy the example, fill in your real values, and then open your own contact page beside it and fix every field where the two disagree. That reconciliation, not the markup itself, is where most of the benefit is.