Most real estate SEO guides tell agents the same thing: add schema markup so Google and ChatGPT understand your site better. That advice is half right. Schema markup does help Google read a page accurately. It is not, on its own, a way to get cited by an AI chat tool, and Google has said so directly.
What schema markup actually is
Schema markup, also called structured data, is a standardized way of stating facts about a page in a format a machine can read directly, instead of a format it has to interpret. Google's own documentation describes it as a way of "providing information about a page and classifying the page content." On a real estate agent's site, that means writing, in code, things like: this is an Organization named [Agent Name], based in [city], reachable at [phone number], offering [service].
The common format is JSON-LD, a small block of structured text placed in the page's code. A visitor never sees it. Search engines and other automated readers do.
What it's good for: removing the guesswork
Without schema, a search engine or an AI crawler has to infer facts about your business from paragraph text, page layout, and context clues. That works most of the time, but it leaves room for error, especially when a brand name is generic or shared with other businesses. Schema removes the guessing. It states the business name, address, phone number, and service area as explicit fields, not prose the reader has to parse correctly.
A few schema types do most of the work on an agent's site. Organization schema is the core entity: business name, address, phone number, logo, and links to verified social profiles. It's the anchor every other page can point back to instead of repeating the same facts on every page. LocalBusiness schema ties a defined service area to the business, which matters when an agent or brokerage works a specific region rather than an entire metro. Review and AggregateRating schema states a rating as structured data instead of leaving it buried in testimonial paragraphs, though it should only ever reflect real, verifiable reviews, not a placeholder number pulled from nowhere. On blog content, BlogPosting or Article schema states the author, publish date, and publisher clearly, so a crawler can attribute the piece correctly instead of guessing.
There is also a type built for this exact business: RealEstateAgent, a schema.org subtype of LocalBusiness. Using the most specific type that truthfully describes the business is good practice, but the specific label matters less than the facts inside it being complete and consistent. An agent site with a modest Organization block whose facts match the visible page and the Google Business Profile is in better shape than one with an elaborate RealEstateAgent block contradicting its own contact page.
What changed with FAQPage schema in 2026
For a few years, FAQ schema was worth adding because a well-marked FAQ section could earn a rich result: an expandable question-and-answer block directly in Google's search results. That changed. As of May 7, 2026, Google stopped showing FAQ rich results in Search entirely, closing out a restriction that had already limited FAQ rich results to a small set of government and health sites since 2023.
FAQPage is still a valid schema type, and Google has said existing markup doesn't need to be ripped out. It just doesn't earn the visual rich result in Google Search anymore. If an agent's site has FAQ content, write it for the reader first. Treat the schema as a bonus for machine readers, not a guaranteed placement in Google's results page.
The claim to be skeptical of: "schema helps you get cited by ChatGPT"
This is where a lot of real estate SEO advice overstates things. Google published its first official guide to optimizing for generative AI features in Search on May 15, 2026, and stated plainly that structured data is not required for AI Overviews or AI Mode, and that there is no special AI-specific markup a site needs to add. The guide's core message: the same things that make a page good for regular Search also make it eligible for Google's AI features, because those features are built on the same ranking and quality systems.
That doesn't mean schema is worthless for AI-era visibility. Crawlers used by other AI systems, retrieval tools, and voice assistants do parse structured data as machine-readable content, and a page with clear, explicit facts is easier for any automated reader to summarize accurately, schema or not. The honest version of the claim is: schema markup helps a machine reader get your facts right. It does not guarantee a citation anywhere, and no current source, from Google or otherwise, confirms it moves citation odds specifically for ChatGPT or Perplexity.
The mistakes that quietly undo the work
Three patterns show up repeatedly on agent sites that technically "have schema." The first is contradiction: the markup says one office address, the contact page shows another, and the Google Business Profile shows a third. A machine reading three versions of a fact trusts none of them fully. This usually happens after a brokerage move, when the visible pages get updated and the markup, which nobody sees, does not.
The second is markup for content that isn't on the page. A rating block describing reviews that live on a portal, or FAQ markup for questions the page never displays, misrepresents the page. Google's structured data guidelines require markup to reflect visible content, and violations risk the site's eligibility for rich results generally, not just on the offending page.
The third is duplication. Some plugin setups emit two or three competing blocks for the same entity (one from the theme, one from an SEO plugin, one pasted in by a past developer), each slightly different. The fix is structural: define the business entity once, give it an @id, and have every other page reference that identifier instead of restating the facts.
Checking for all three takes minutes and no budget. Paste the homepage into validator.schema.org or Google's Rich Results Test and read what comes back: how many blocks, which types, and whether the facts match what the page visibly says.
Where the real gain is
If schema isn't a guaranteed lever for AI citation, what is worth doing? The same thing that has always mattered: making sure a page actually states its facts clearly in visible text, not just buried in a JSON-LD block. Schema is a supplement to clear content, not a substitute for it. A page with a vague "About" paragraph and a perfect Organization schema block is still a vague page to a human reader and to most AI systems, which read the visible content of a page, not just its structured data.
Saige's own site uses this layered approach: an Organization entity
anchors the core business facts once, and every other page (the homepage,
about page, contact page, service pages) links back to that same entity
through its @id reference instead of repeating the fields. That keeps the
facts consistent everywhere a crawler looks, which is the actual goal:
consistency and clarity, not a specific markup type chasing a specific AI
platform. Sites rebuilt by Saige get the same structure generated
automatically, so the markup stays in step with the visible content instead
of drifting away from it over the years.
The takeaway
Add schema markup because it helps Google, and any other automated reader, understand your business accurately, not because it guarantees a citation in an AI chat tool. Keep the visible page content clear and specific first. Schema is the machine-readable backup for facts a human reader can already find on the page, not a shortcut around writing them.
Related: Getting described the same way everywhere · How AI search is changing real estate marketing · Does a website redesign hurt your Google rankings · What is AEO for real estate agents
