GEO for Real Estate Agents and Brokerages: Getting Named When Buyers Ask AI

Commission questions, school districts and the cash-offer letter

Real estate questions come in three moods. Buyers research for months. Sellers agonize over one decision. And since the NAR settlement practice changes took effect on August 17, 2024, almost everyone is confused about how agents get paid: offers of compensation can no longer appear on MLSs, and agents must have a written agreement with buyers before touring a home. Confused people ask machines. These are the kinds of questions we test for agents and brokerages:

• "Best real estate agent in San Antonio for first-time buyers."
• "How much do realtors charge in Texas after the NAR settlement?"
• "Homes for sale near good schools in Alamo Heights."
• "Should I sell my house for cash or list it?"
• "Do I have to sign an agreement with an agent before I tour a house?"
• "Real estate agent near me who speaks Spanish."

Look at what an engine needs to answer these well. A fee question needs a page that explains, in numbers you choose to publish, how you are paid. A neighborhood question needs a page about the neighborhood. A who-should-I-call question needs a license record, reviews and listings that agree. Agents who publish only a headshot and an IDX search box give the engine a portal to cite instead of a person. Google's AI Overviews and AI Mode both draw on the same Search index your site already lives in.

The portal owns the listing; nobody owns the agent recommendation yet

Zillow's app sends ChatGPT users back to Zillow to schedule tours, connect with an agent or explore financing. That covers the house. The question of which human to trust is answered from a different pile of sources, most of which you can influence.

Teams add a wrinkle. An agent, a team and a brokerage are three entities, and engines muddle them when the names drift. TREC Rule 535.155 already forbids team names that use words like "brokerage," "company" or "associates" to imply a separate firm, which helps: one broker name, one team name, one spelling, everywhere. The broader cleanup method is in entity optimization.

Fair housing, TREC's homepage links and the broker's name in every ad

Real estate advertising runs under federal and state rules that land squarely on the pages AI engines read. These are the ones every Texas brokerage site has to account for.

• Fair Housing Act section 804(c) (42 USC 3604(c)) makes it unlawful to make, print or publish any notice, statement or advertisement for the sale or rental of a dwelling that indicates a preference, limitation or discrimination based on race, color, religion, sex, handicap, familial status or national origin.
• HUD's 2024 digital advertising guidance is withdrawn. According to a law firm's summary of the Federal Register notice of April 6, 2026, HUD withdrew its April 29, 2024 guidance on the Fair Housing Act and digital ad platforms, effective September 17, 2025. The Act and HUD's regulations still apply in full.
• TREC Rule 531.20: brokers and sales agents link the Information About Brokerage Services form from a readily noticeable place on the homepage, labeled "Texas Real Estate Commission Information About Brokerage Services" (at least 10-point type) or "TREC Information About Brokerage Services" (at least 12-point).
• TREC Rule 531.18: each license holder provides the Consumer Protection Notice (brokers also display it in each place of business they maintain) and links it from the homepage, labeled "Texas Real Estate Commission Consumer Protection Notice" (10-point) or "TREC Consumer Protection Notice" (12-point); on social media the link may sit on the profile.
• TREC Rule 535.155: every ad names the license holder or team placing it, and the broker's name must be at least half the size of the largest contact information for any agent or team. On social media the required details may be on the profile or a directly linked page.

The GEO consequence is specific. Neighborhood pages are the pages engines quote most for agents, and they are also where fair housing risk lives. Our practice is to write about the place and the property: housing stock, commute routes, parks, the name of the school district with a link to it, HOA details. We do not write about who the buyer ought to be. An engine paraphrasing a page that describes people is paraphrasing the wrong thing.

Neighborhood guides, fee pages and bios an engine will lift

Engines prefer pages that answer a question in their first lines and back it with facts. For a brokerage that means fewer slogans and more of the following.

In markup, schema.org's RealEstateAgent is a LocalBusiness subtype; pair it with Organization markup on the homepage and keep both identical to the visible text. Google says no special markup is needed for its AI features, so this is tidiness, not a trick. Check your listing search, too: if listings load by script or inside a frame, look at the raw HTML, because many AI crawlers do not run JavaScript. JavaScript and AI crawlers shows the ten-minute test.

Reviews from clients who move once a decade

A restaurant hears from guests every night. An agent hears from a client at closing and then, perhaps, in eight years. That makes every closing the one chance to ask, and every review heavy. Google lists review count and rating among the things that can help local ranking, and an engine describing you will lean on whatever volume exists. The rules for asking are the same ones every business faces, with one real estate trap: the closing gift.

In Athena CMS, which comes with every site we build, the Echo agent drafts review replies for your approval and places review quotes on the matching neighborhood pages (how Athena works). The longer explanation is in reviews and AI recommendations, and if an answer already quotes a sour client, see AI quotes a bad review.

The first thirty days for an agent or a brokerage

1. 1

   Record what the engines say now

   Ask ChatGPT, Gemini, AI Mode, Perplexity and Copilot ten questions a client would ask, naming neighborhoods. Our free AI visibility check does this live through each engine's API and lists every business named (the consumer apps may answer a little differently).

2. 2

   Match everything to TREC

   Agent and broker names, license numbers and the sponsoring broker, identical on the site, Business Profile, Apple Business, Bing Places, Yelp and MLS.

3. 3

   Fix the homepage requirements

   Confirm the IABS and Consumer Protection Notice links carry the exact labels and type sizes TREC requires.

4. 4

   Publish the fee page

   Explain the buyer agreement and how you are paid, in plain words, approved by your broker.

5. 5

   Write two neighborhood guides

   Pick the two areas you sell most. Describe the place, never the people, and date the page.

6. 6

   Start asking at every closing

   Same request, every client, no gift attached.

7. 7

   Compare on day thirty

   Put the same ten questions to the engines again and note which sources each one cited this time.

Listings usually settle within weeks; answers shift more slowly and unevenly. If a rival agent gets named and you do not, ChatGPT recommends a competitor lays out the usual causes.

How DataCram helps brokerages, and when a solo agent should go it alone

There is no separate GEO invoice: the work sits inside a monthly SEO program with a defined scope and no long contract. Programs begin at $625 a month, and most local businesses spend between $625 and $2,250 (see the SEO pricing page). A free GEO audit sets the tier, based on how contested your neighborhoods are and how much the site and listings need. GEO pricing covers the levers; financing exists.

The program tracks your clients' questions across the engines with AICiteBeam, cleans up the entity record from TREC to Bing Places, runs review requests and replies, writes neighborhood and process pages inside your broker's review, and reports the actual answers monthly. If the site is the problem, our real estate industry page covers how we build brokerage websites.

The honest answer for a solo agent in a small market: you can do most of this yourself. Match your name to TREC, claim the listings, write one honest fee page and two neighborhood guides, ask every client at closing. A brokerage with several teams and a dozen neighborhoods has more to keep consistent, which is where a program earns its fee. Related trades face related questions: see GEO for mortgage lenders, insurance agencies and law firms, or browse every industry guide.