A real estate website usually has three kinds of page, and each one gets its own Schema.org description. Mark the brokerage or office page with RealEstateAgent, each agent profile with Person connected to the brokerage, and each property page with RealEstateListing, a property type such as House or Apartment, and an Offer for the sale or rental terms.

Add BreadcrumbList wherever a visible breadcrumb trail appears. Every price, bedroom count, address and agent detail in the markup also has to appear on the page, and listing wording deserves a fair-housing check before it goes live.

The JSON-LD examples use a fictional brokerage on example.com. Google's documented features for a site like this are Local Business and Breadcrumb, so the listing types are best treated as an accurate description of the page, not a route to a special result.

Key takeaways

  1. One type per pageUse RealEstateAgent for the brokerage, Person for each agent, and RealEstateListing with a property type and Offer for each listing page.
  2. Connect with @idGive the brokerage, agents, listings and offers stable @id values, and point worksFor, affiliation and offeredBy at them so the entities link up.
  3. Google features that applyGoogle's Local Business and Breadcrumb features apply here. Listing and property types are Schema.org vocabulary with no Google feature requirements attached.
  4. Match the visible pageEvery price, bedroom count, address and agent detail in the markup must appear on the page, and both change together.
  5. Fair-housing wordingDescribe objective property facts, not who a home suits, and ask a qualified housing professional about uncertain wording.
  6. Test after publishingRun the Schema Markup Validator and Rich Results Test, then inspect the live page in Google Search Console and Bing Webmaster Tools.

Which type belongs on which page

The page decides the type, so start by naming the page. Four kinds of page cover most real estate sites.

  • The brokerage or office page uses RealEstateAgent, the Schema.org subtype for a real estate agent or agency.
  • An agent profile uses Person, tied to the brokerage with worksFor or affiliation.
  • A property page uses RealEstateListing for the page, a property type such as House, Apartment or SingleFamilyResidence for the home, and Offer for the terms.
  • Any page with a visible breadcrumb trail uses BreadcrumbList.

A brokerage can choose between RealEstateAgent and plain Organization. In Schema.org, RealEstateAgent is a subtype of both LocalBusiness and Organization, so it keeps everything an organization can say and adds what kind of business it is. Use it when the page visibly describes the real estate business. The organization schema guide covers the general case.

A business that sells something other than property has its own pages. See the guides for law firms, restaurants, medical practices, home services and ecommerce stores.

Mark up the brokerage with RealEstateAgent

Put the brokerage block on the home page or the office page, whichever visibly describes the business. The properties to fill in are the name, address, telephone, URL, logo, hours and service area, and each one should match what a visitor reads on the page.

Google's Local Business documentation lists name and address as required for its Local Business feature. It recommends further fields such as telephone, URL, logo, opening hours and geo-coordinates. Because RealEstateAgent is a Local Business subtype, the same fields apply.

json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "RealEstateAgent",
      "@id": "https://example.com/#brokerage",
      "name": "Harborline Realty",
      "url": "https://example.com/",
      "logo": "https://example.com/images/harborline-logo.png",
      "telephone": "+1-555-010-0142",
      "address": {
        "@type": "PostalAddress",
        "streetAddress": "100 Example Avenue",
        "addressLocality": "Example City",
        "addressRegion": "IL",
        "postalCode": "60601",
        "addressCountry": "US"
      },
      "openingHoursSpecification": {
        "@type": "OpeningHoursSpecification",
        "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
        "opens": "09:00",
        "closes": "17:00"
      },
      "areaServed": {
        "@type": "AdministrativeArea",
        "name": "Example County"
      }
    }
  ]
}

Replace Harborline Realty, the address, the telephone number, the hours and the county with your own details, and drop any property the page does not show. The @id is a stable identifier for the brokerage. Give it a fragment on your home page URL and reuse the same value every time another block refers to the brokerage.

The LocalBusiness schema generator builds this kind of block from a form, and the local business schema guide explains each field. A brokerage with several offices can follow the multiple locations guide.

Connect each agent with Person

An agent profile page describes one human being, so it uses Person. Fill in only what the profile shows. Schema.org's Person type offers these properties for an agent:

  • name, jobTitle, image and url
  • telephone, and email only when the page displays it
  • worksFor or affiliation, which point to the brokerage
  • sameAs, for professional profile links that identify the same person
  • knowsAbout and knowsLanguage

Use worksFor when the agent is employed by the brokerage. Use affiliation when the relationship is broader and the person is associated with the organization. Either way, point to the brokerage by its @id so the two descriptions join up.

json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Person",
      "@id": "https://example.com/agents/jordan-lee#person",
      "name": "Jordan Lee",
      "url": "https://example.com/agents/jordan-lee",
      "image": "https://example.com/images/jordan-lee.jpg",
      "jobTitle": "Real Estate Agent",
      "worksFor": {
        "@id": "https://example.com/#brokerage"
      },
      "knowsLanguage": ["en", "es"]
    }
  ]
}

Swap in the agent's own name, profile URL, photo and job title. The knowsLanguage values are language codes, so list only the languages the agent actually works in and the profile mentions. Google has no dedicated rich result requirement for a Person used this way, but the node still gives Bing and other systems that read Schema.org a clear description of who the agent is.

The person and author schema guide goes further, and the Person schema generator builds the block. For sameAs, see entity SEO with sameAs.

Describe a listing with RealEstateListing, a property type and Offer

A listing page needs three connected descriptions. RealEstateListing describes the page itself and can carry datePosted, name, url, image, description and a link to the offer. A specific Accommodation subtype such as House, Apartment or SingleFamilyResidence describes the physical home. Offer carries the terms: price, priceCurrency, availability, businessFunction and url.

Property facts belong on the property node. PostalAddress, GeoCoordinates, floorSize, numberOfBedrooms, numberOfBathroomsTotal, yearBuilt and amenityFeature all fit there, as long as the facts are visible on the page and accurate. Schema.org's Accommodation vocabulary also covers pets, occupancy and permitted usage, and RealEstateListing defines the page-level type.

A listing page card linked by straight lines to a property card with house, bed and bath icons and to a separate tag tile with a coin icon
Figure 1. A listing page, the property it describes and the terms of the offer are three connected descriptions.
json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "RealEstateListing",
      "@id": "https://example.com/listings/example-house#listing",
      "url": "https://example.com/listings/example-house",
      "name": "Example House in Example City",
      "description": "Fictional three-bedroom house with a two-car garage and a fenced yard.",
      "datePosted": "2026-10-15",
      "image": [
        "https://example.com/images/example-house-front.jpg",
        "https://example.com/images/example-house-kitchen.jpg"
      ],
      "mainEntity": { "@id": "https://example.com/listings/example-house#property" },
      "offers": { "@id": "https://example.com/listings/example-house#offer" }
    },
    {
      "@type": "House",
      "@id": "https://example.com/listings/example-house#property",
      "name": "Example House",
      "url": "https://example.com/listings/example-house",
      "address": {
        "@type": "PostalAddress",
        "streetAddress": "200 Sample Street",
        "addressLocality": "Example City",
        "addressRegion": "IL",
        "postalCode": "60602",
        "addressCountry": "US"
      },
      "numberOfBedrooms": 3,
      "numberOfBathroomsTotal": 2,
      "floorSize": { "@type": "QuantitativeValue", "value": 1800, "unitCode": "FTK" },
      "yearBuilt": 2018
    },
    {
      "@type": "Offer",
      "@id": "https://example.com/listings/example-house#offer",
      "url": "https://example.com/listings/example-house",
      "businessFunction": "https://schema.org/Sell",
      "price": 425000,
      "priceCurrency": "USD",
      "availability": "https://schema.org/InStock",
      "itemOffered": { "@id": "https://example.com/listings/example-house#property" },
      "offeredBy": { "@id": "https://example.com/agents/jordan-lee#person" }
    }
  ]
}

Every @id in this block is a label you choose, and the references between them are what join the page, the home and the terms. Replace the fictional address, room counts, floor area and year with the figures shown on your listing page, change House to Apartment or another subtype when that fits the home, and set the price and currency to the amounts a visitor sees. The unitCode value FTK stands for square feet.

The offeredBy reference on the offer points to the listing agent's @id. Keep each page's block readable on its own: if the listing page does not also define the agent, include the agent's name and @id in that page's graph so the reference has something to resolve to. The JSON-LD graph guide explains how @id links work.

Add BreadcrumbList to pages with a visible trail

A breadcrumb trail such as Home, Listings, then the property name is worth marking up when visitors can see it. Google's Breadcrumb documentation requires a BreadcrumbList with at least two ListItem entries. The itemListElement property is required for that feature, each ListItem needs position and name, and item is required except for the final crumb when the page URL can be inferred.

json
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://example.com/" },
    { "@type": "ListItem", "position": 2, "name": "Listings", "item": "https://example.com/listings/" },
    { "@type": "ListItem", "position": 3, "name": "Example House in Example City" }
  ]
}

Match the names and order to the trail on your page. The breadcrumb schema guide and the breadcrumb schema generator cover longer trails and category paths.

Choose properties by page purpose

Which properties belong where depends on the page and on what each consumer documents. Google's Local Business feature has required fields, the Breadcrumb feature has its own, and the listing and property types are Schema.org vocabulary with no Google feature requirements attached. Label listing properties as Schema.org properties, never as Google requirements.

Properties by page type

Hover over or tap a row to highlight it. On small screens the table scrolls sideways.

TypeProperties to includeGoogle feature status
RealEstateAgentname, address, telephone, url, logo, openingHoursSpecification, areaServedGoogle Local Business: name and address required; telephone, URL, logo and hours recommended.
Personname, jobTitle, image, url, telephone, worksFor, sameAs, knowsLanguageNo dedicated Google rich result requirement for this use.
RealEstateListingname, url, image, description, datePosted and a link to the offerSchema.org vocabulary properties, not Google requirements.
House or Apartmentaddress, numberOfBedrooms, numberOfBathroomsTotal, floorSize, yearBuilt, amenityFeatureSchema.org vocabulary properties, not Google requirements.
Offerprice, priceCurrency, availability, businessFunction, url, itemOfferedSchema.org properties with no Google real estate feature requirements attached.
BreadcrumbListitemListElement with position, name and item on each ListItemGoogle Breadcrumb: itemListElement, position and name required; item required except on the final crumb.

What Google shows and what the types still do

Google's structured data gallery lists features such as Local Business, Organization, Breadcrumb, Product, Event, Article and Vacation Rental. For a real estate site, Local Business applies to the brokerage and Breadcrumb applies to trails. RealEstateListing, House, Apartment and Accommodation are valid Schema.org vocabulary, and Google's search features for a site like this are Local Business and Breadcrumb, so a listing page should not be expected to earn a dedicated listing result.

Google says structured data helps it understand page content, that display is discretionary, and that valid markup does not guarantee a rich result or a ranking improvement (general structured data guidelines). It also says its generative Search features need no special Schema.org markup, while it continues to recommend structured data for eligible Search features (AI features and your website).

Support for features changes, so read the dates. Google's documentation updates record that the FAQ rich result stopped appearing in Search on May 7, 2026, and that the FAQ documentation was removed in June 2026. FAQPage remains a valid Schema.org type, and FAQ content still helps readers, but it is not a current Google FAQ rich result opportunity. The rich results guide explains the difference between valid markup and a display.

Make the markup readable for Bing, Yandex and Apple

Bing says it supports Schema.org in JSON-LD, Microdata and RDFa, and its structured data documentation covers businesses and organizations, people, products and offers, which matches the brokerage, agent and offer nodes above. Yandex documents JSON-LD inside a script tag and a validator, and not every Schema.org type is supported by its services (Yandex Schema.org usage).

About Applebot explains that Applebot crawls public web pages for Apple Search, Siri and Spotlight, and that pages marked noindex do not appear in Spotlight or Siri Suggestions. For DuckDuckGo, keep the page crawlable and inspect it in Bing Webmaster Tools, because Bing results may be used by other services. See whether AI systems use schema markup.

Keep the markup and the listing wording accurate

Google's guidelines tie structured data to visible content. Every price, bedroom count, address and agent detail in the JSON-LD should appear on the page a visitor reads, and the two should change together. Wix says the same of its manual markup, which must correspond to visible page content. Leave out personal data, amenities that do not exist, unverified neighborhood claims and hidden text.

A useful habit is to read the printed listing next to the markup, line by line, before publishing.

A woman with grey-streaked dark hair holds a pen over a printed listing sheet beside an open laptop on a desk in a small real estate office
Figure 2. Reading the printed listing next to the markup shows whether the price, rooms and address match.

Fair-housing care applies to the words on the page, and structured data does not make wording acceptable that would not be acceptable in the listing. HUD's advertising rules in 24 CFR Part 109 identify words, phrases and symbols that may convey discriminatory preferences or limitations, and the Fair Housing Act names the protected classes. Describe the home through objective facts, such as:

  • bedroom and bathroom count, and floor area
  • parking, accessibility features and building features
  • transit distance and nearby public facilities
  • sale or lease terms, and the pet policy

Wording about who a home suits, such as "perfect for families," "ideal for Christians" or "adult community," and unverified claims such as "safe neighborhood" or racial and ethnic descriptors, are the kind of language to put in front of a qualified housing professional first. State and local law can add requirements beyond the federal rules, so have a qualified local housing or legal professional look at the wording whenever it is uncertain.

Add the markup on your platform

Check what your platform already outputs before adding a second description of the same business. On WordPress, use the site's existing SEO or schema system when it already emits Organization, WebSite and Breadcrumb markup, and add custom listing and agent data only where the plugin supports it, so no node appears twice. Plugin behavior changes by developer and version, so read the rendered HTML, not just the editor. The WordPress guide and the duplicate markup guide help here.

Shopify's structured_data Liquid filter applies to product and article objects, so listing markup needs a custom theme or app implementation, and only when the listing content is visible and kept in sync. Wix supports custom JSON-LD through page SEO settings. It must be valid, must match visible content and is limited to 7,000 characters, with up to five markups per page in the manual workflow. See the Wix and Shopify guides.

Squarespace automatically emits structured data for blog posts, events, local businesses, organizations, products and websites, and says those built-in schemas cannot be edited or removed through standard controls. Read the rendered page before adding custom JSON-LD so the brokerage is not described twice. The Squarespace guide has the steps.

Webflow offers a page-level Schema markup field for manual JSON-LD, and Collection pages can bind fields such as name, address, price and image. Keep one implementation path per page. On Next.js, the JSON-LD guide recommends a script tag in layout.js or page.js and replacing < with \u003c when serializing with JSON.stringify. On HubSpot, use head injection or a developer-managed template, then confirm the published page contains the intended JSON-LD. The Webflow, Next.js and React and HubSpot guides go deeper.

Test the markup across search systems

Test the rendered page, not the template. The Schema Markup Validator checks vocabulary validity, and Google's Rich Results Test checks eligibility for the features Google supports, which here means Local Business and Breadcrumb. Neither tool guarantees that anything will be displayed. The Schema Nova validator can check a pasted block, and the Rich Results Test comparison explains when to use which.

After publishing, inspect the live page in Google Search Console and in Bing Webmaster Tools. The URL Inspection tool in Bing Webmaster Tools includes a Markup card that can inspect JSON-LD and other recognized markup. Run Yandex's validator when Yandex visibility matters to you. Update the markup whenever the visible listing changes.

Setup order for real estate markup

  1. Name the pageDecide whether it is the brokerage, an agent profile, a listing or an article, then pick the most specific type.
  2. Write visible factsAdd only facts the page shows, and connect the entities with stable @id values.
  3. Add the JSON-LDUse the platform's supported field or template, and avoid duplicating output the platform already emits.
  4. Validate the codeCheck JSON syntax and Schema.org vocabulary with the Schema Markup Validator.
  5. Inspect the live pageRun the Rich Results Test for Local Business and Breadcrumb, then inspect in Google Search Console and Bing Webmaster Tools.
  6. Keep it currentUpdate the markup whenever the visible listing changes, including price and availability.

Keep listing pages open to AI search crawlers

OpenAI says a website must allow OAI-SearchBot to crawl it to be eligible for inclusion in ChatGPT search, and that search placement is not guaranteed. Its crawler documentation separates OAI-SearchBot, used for search, from GPTBot, used for possible training use. Perplexity describes web search as retrieving, reading and synthesizing sources.

Google says its AI Overviews and AI Mode have no additional technical requirements beyond normal Google Search eligibility. For Bing-powered experiences, use Bing's structured data handling and Bing Webmaster inspection. No markup guarantees citations, so keep public listing pages crawlable, with complete and current property facts in the visible text. The AI search guide covers the wider picture.

Questions people ask

Does Google guarantee a rich result once the markup validates?

No. Google says valid markup does not guarantee a rich result or a ranking improvement, and display is at its discretion. A passing test confirms parsing or feature eligibility, which is a different thing from being shown.

How should listing markup handle a changed price?

Change the visible price and the price value in the Offer together, and update availability when the status changes. Markup that disagrees with the page conflicts with Google's rule that structured data describe visible content.

Bing says it supports Schema.org JSON-LD and uses structured data to understand page content. Yandex documents JSON-LD and a validator, and says not every Schema.org type is supported. For ChatGPT search, OpenAI says eligibility depends on allowing OAI-SearchBot, with placement not guaranteed.

What fair-housing issues apply to listing descriptions?

HUD's advertising rules identify words, phrases and symbols that may convey discriminatory preferences or limitations, and state and local law can add rules. Describe objective property facts, and ask a qualified housing professional about wording you are unsure of.

Quick check

Choose an answer to see feedback. Nothing is scored or saved.

Question 1 of 3

Which type describes the physical home on a listing page?

Question 2 of 3

Which property ties an employed agent's Person node to the brokerage?

Question 3 of 3

Can the markup contain a price that is not visible on the page?

Glossary

Select a term to see its definition.

RealEstateAgent. A Schema.org type for a real estate agent or agency, and a subtype of LocalBusiness and Organization.