AEO for Commercial Real Estate: How Brokerages, Owners and Developers Get Named in AI Answers

Updated October 2026 · Written and maintained by the Progression Agency strategy team

AEO for commercial real estate is answer engine optimization for brokerages, landlords, owners and developers: the work of making a firm’s brokers, buildings, markets and closed deals readable and checkable, so that ChatGPT, Claude, Gemini, Perplexity, Microsoft Copilot, Google AI Overviews and AI Mode can name the firm when a tenant, an investor or a site selector asks a commercial property question. It starts from a feature of this market: much of what matters sits behind a login, a confidentiality agreement or a bot defense, while the facts an outsider can confirm sit in license lookups, designation directories, securities filings, county records and economic development databases. It is written for office, industrial, retail, multifamily and medical office firms whose next client may begin with a question typed to a machine. Progression Agency is based in New York City and works with clients across the United States and worldwide.

What is AEO for commercial real estate?

It is the work of turning a commercial real estate firm into material an AI answer can safely use. That takes three conditions. The firm’s facts exist as text on pages an assistant is permitted to fetch. Each fact can be checked against a record the firm does not control. And each page says it plainly enough that it can be repeated with the firm’s name on it. The general method of answer engine optimization applies to every industry; this page covers what commercial property adds to it.

It is a different job from ranking. Commercial real estate SEO on this site covers search results: market and submarket pages built to rank, listing pages and what becomes of them after closing, local profiles for each office. This page does not go over that ground again. It begins with another reader, a language model assembling one answer out of several sources, and asks what that reader is able to see of a business that keeps its best information private.

Who this page is written for

• Tenant representation brokers and occupier advisory teams whose clients arrive having already asked an assistant what space costs.
• Landlord leasing and investment sales teams that want to be on the shortlist an owner is handed.
• Owners and landlords of office, industrial, retail, multifamily and medical office buildings with space to fill.
• Developers whose projects are discussed by lenders, neighbors, tenants and journalists before they open.
• Industrial park and land owners whose sites are screened by site selectors long before anyone calls.
• Sponsors and operators raising capital, who have to decide what may be said in public at all.
• Marketing leads at any of these who have been asked, for the first time, what the AI says about the firm.

What this page leaves to others

Residential agents and teams are covered in AEO for real estate agents, third-party managers in AEO for property management and apartment communities in AEO for apartments. The wider commercial program has pages of its own: commercial real estate marketing, commercial real estate lead generation and commercial real estate website design. What follows is only about being read, checked and cited by machines.

How do tenants, investors and site selectors ask AI assistants about commercial property?

With a size, a place, a property type and a deadline in one request, and usually with a request for names. Each kind of asker expects a different kind of proof, and each answer is assembled from a different kind of page. The wording below is illustrative; the pattern is what matters.

Tenants and occupiers

A business owner looking for space rarely knows the vocabulary. The question arrives as a headcount, a place and a budget, and the answer the assistant gives sets the expectation a broker later has to meet or correct. Property types change the follow-up questions: an office tenant asks about cost and commute, a medical practice about build-out and whether to own, a retailer about co-tenancy and the suites coming free, a distribution occupier about specification and dates. Two things decide whether a brokerage is part of the answer: a page that states what space of that kind costs in that submarket, with a date on it, and a named person whose specialty matches the request.

Owners and landlords

An owner asking for the best broker for a building is asking for a shortlist, and an assistant builds a shortlist from whatever distinguishes one firm from another in text. Assignments described only in a pitch deck do not exist for this purpose. Assignments written up on the firm’s site, by property type and submarket, with the status and the date, do.

Investors and sponsors

Investors use assistants as a first screen: which markets, which property types, which operators. Then they check. A sponsor whose legal entity names, principals and offering history line up across its website, its securities filings and state records passes that screen quickly. One whose website uses a brand name that appears in no filing leaves the assistant, and the investor, with a gap to fill by guessing.

Site selectors and corporate real estate teams

Site selection consultants and corporate real estate teams work for a company choosing a location, often confidentially and often before any broker is engaged. Their early questions are about places, not buildings: labor, power, rail, highway access, sites of a given size. An assistant answers them from economic development organizations, federal statistics and whatever owners and developers have published about their sites. An industrial park or a vacant building is part of that answer only if its acreage, zoning, utilities and access exist as text somewhere an assistant can read, which, as the next section shows, is not guaranteed by being in a state database.

Lenders, appraisers and service firms

These readers ask factual questions about a property or a person, and they are the likeliest to notice a wrong answer. They are also where referrals come from. A firm whose published transaction facts agree with the recorded deed is one they can rely on, and a firm that sells their kind of service, from debt placement to property tax appeals, is asked about in the same way a broker is. AEO for mortgage brokers and AEO for financial advisors cover two neighboring professions.

Which sources do AI assistants draw on for commercial real estate answers?

Six kinds, and a firm controls only one of them completely. Listing marketplaces, brokerage websites and broker bios, economic development organizations, business and trade press, public registers and government statistics each supply part of a commercial answer. How much of each an assistant can actually read varies more in this industry than in most, so we checked.

Listing marketplaces

LoopNet presents itself as a marketplace for commercial property for sale and for lease, and Crexi calls itself the commercial real estate exchange. CoStar sells research by subscription: an Internet Archive copy of its homepage from August 31, 2026 sends users to a login and tells everyone else that some of its news stories are intended only for CoStar subscribers. What a firm controls on each platform is covered on our commercial real estate marketing page. For AEO the question is narrower, and a later section on this page reports what could and could not be confirmed about whether an AI crawler may read the platforms at all.

Brokerage websites and broker bios

The firm’s own site is the one source in the table whose access rules, wording and dates are entirely the firm’s decision, and a broker’s bio is the page an assistant reaches for when a person is named. Commercial real estate SEO sets out what a profile should contain for search. For AEO the test is agreement: the name, license number, designations and specialty on the bio should match the license lookup, the designation directory and the way the trade press has described the same person.

Economic development organizations and site databases

State and regional economic development organizations answer site questions for a living, and many publish databases of available sites and buildings. Ohio shows how such a database can be closed to most AI crawlers. JobsOhio, which describes itself as the state’s private, nonprofit economic development organization, points its Available Ohio Sites button at a property search hosted on ZoomProspector, a mapping product sold to economic developers for publishing available commercial properties. On October 5, 2026 that host’s robots.txt admitted Googlebot and Bingbot and refused every other crawler, which takes in OAI-SearchBot, Claude-SearchBot and PerplexityBot, and the search page’s HTML held filters and no property records.

An owner whose building appears only in a database like that is reachable through Google and Bing and not through those three crawlers. The remedy is the same as for marketplaces: a page of the owner’s own for the building or site, carrying the same acreage, square footage, zoning, utilities and contact as the database entry, and a request to the economic development organization to link to it. Certification programs, such as the SiteOhio Authenticated designation that JobsOhio announces for qualifying parks, give a site a third-party status that fits in one sentence, which is exactly the kind of claim an assistant can repeat. JobsOhio’s site selection page shows what a state program publishes. Nationally, the International Economic Development Council is a nonprofit membership organization of more than 4,500 economic developers, and SelectUSA is the Department of Commerce program that facilitates business investment into the United States.

Business and trade press

Coverage matters because it is one of the few places where a firm’s deal is described by somebody else. We asked eight business and trade publications that report commercial property deals for their robots.txt file on October 5, 2026.

Where the file could be read, nothing in it stood between OpenAI’s, Anthropic’s or Perplexity’s search crawlers and the articles. That makes the press a channel worth using carefully. Publish the deal page on the firm’s site before the announcement goes out, with the size, the property type, the month and the firm’s role, so that the article and the page say the same thing on the same day. An assistant that finds two accounts of one deal that disagree has to choose, and it may not choose yours. Digital PR covers how coverage is earned.

Registers and statistics

The last two rows of the table are where an assistant, or a careful reader, goes to check. They are covered in the next two sections, because in commercial real estate the check is often more important than the claim.

What can an AI assistant verify about a commercial real estate firm, and what can it not see?

It can verify what is public and has an address of its own, and nothing else. A license record, a designation listing, a securities filing, a recorded deed and a page of text on an open website can be read by anyone. A subscription database, a document behind a confidentiality agreement and a figure that lives only in a broker’s head cannot.

What sits behind a wall

Commercial real estate runs on restricted information. Subscription research sits behind a login, and offering memoranda are commonly released only after a confidentiality agreement is signed. Google’s documentation puts the consequence plainly from the other direction: robots.txt is not the way to keep a page out of Google, and the ways that work are a noindex rule or a password. Whatever a firm has placed behind a login for sound commercial reasons is, for an assistant, simply not there.

A lookup is a form, not a page

Most license lookups are search forms: a visitor types a name or a number and the record is produced on request. The Association of Real Estate License Law Officials runs a multi-state database drawing on 43 participating jurisdictions, four of them Canadian provinces, and puts a captcha in front of its search; New York, Illinois, Pennsylvania and Ohio were not on its list of jurisdictions on October 5, 2026. Anthropic states that its crawlers will not attempt to bypass captchas. So the record exists and is public, yet a crawler will not come across it. Some regulators do give each record an address of its own. California’s Department of Real Estate is one: a record there opens directly from its license ID and shows the license type, status and expiration date. Where a regulator works that way, link the record from the broker’s profile. Where it does not, the license number in plain text is what lets a reader, or an assistant browsing on one person’s behalf, run the check.

Which public registers hold checkable commercial real estate facts?

Five families of record: state license lookups, designation directories, securities filings, county land records and government statistics. None of them is a marketing channel. Each is a place where a claim on a firm’s website either agrees with an outside record or does not.

License and designation records

California, Texas, Florida and New York each run a free lookup: California’s Public License Lookup, the Texas Real Estate Commission’s License Holder Search, Florida’s licensee search and New York’s public license search. Florida searches by name, license number, city or county and license type; New York by name, license number, city or county. For designations, Find a CCIM filters designees by location, profession and expertise, and SIOR’s member directory searches by name or company. ARELLO’s database covers the participating states in one search.

The designations are worth stating in words because of what stands behind them. The CCIM Institute says more than 20,000 professionals have completed its designation program since 1967; the education component is four designation courses, an online ethics course, negotiation training and electives, followed by a portfolio of qualifying experience and a full-day comprehensive exam. SIOR describes itself as having 4,000 members in over 50 countries. A designation is a claim somebody else has already vetted, which is exactly the kind of claim an assistant can lean on.

Securities filings: EDGAR for REITs and Form D notices

The SEC’s EDGAR database gives free public access to the filings companies make, and the Commission’s investor education site explains how to use EDGAR to research them. Two parts of it matter to commercial property. The SEC’s investor site says many REITs are registered with it, and that EDGAR holds a registered REIT’s annual and quarterly reports and any offering prospectus, so anything a REIT’s website says about its portfolio can be set beside them. And a company that sells securities under Regulation D files a Form D after its first sale: a brief notice giving the names and addresses of its executive officers, the size of the offering and the date of first sale. The SEC’s own advice to investors is to look a company up on EDGAR to see whether it has filed a Form D. A sponsor should assume that an assistant asked about it can be pointed at the same notice, and should make sure the names match.

County records: deeds, mortgages and assessments

Land records are kept locally, so there is no single national search. New York City’s ACRIS is one example: the city’s Department of Finance says it lets anyone search property records and view document images for Manhattan, Queens, the Bronx and Brooklyn from 1966 to the present. Assessors work in a similar way; Cook County’s, for instance, offers a search by address, PIN or neighborhood. For a firm the use is defensive as much as promotional. When a transaction page says a building sold in March and the recorded deed says May, the page is the one that looks wrong.

Government data for market pages

A market page that offers only the firm’s opinion gives a reader nothing to test. Federal series supply the testable part. The Census Bureau’s County Business Patterns counts establishments, employment and payroll by industry every year, though its newest data, as of October 2026, describe 2023. The Bureau of Labor Statistics publishes the Quarterly Census of Employment and Wages by county, metropolitan area and state, and the Bureau of Economic Analysis publishes gross domestic product by county. The Federal Reserve’s Beige Book gives a dated, district-by-district reading eight times a year. One caution: the Census Bureau’s Building Permits Survey covers new privately owned residential construction, so it speaks to multifamily supply and says nothing about offices or warehouses. State the year of every federal figure in the sentence that uses it, because an assistant quoting the sentence will not see the footnote.

What do the commercial listing platforms let AI crawlers read?

From outside, mostly it cannot be said, and that is the finding. On October 5, 2026 we requested the robots.txt file of eleven commercial listing and property-data sites with a plain command-line request. Seven refused it. Four returned the file, and none of those four named an AI crawler.

A refusal of our request is not a statement about what Googlebot or an AI crawler receives; a site can treat a search engine’s crawler differently from an unidentified script. It does mean the rules cannot be confirmed from outside, so nobody should tell a brokerage with confidence that its marketplace listings are, or are not, open to ChatGPT or Perplexity. The Internet Archive holds copies of six of the seven refused files, captured between December 2025 and May 2026, and none of them named an AI crawler either. The April 2026 copy of LoopNet’s file did close several paths to every crawler, among them its /broker/ and /market/ sections.

What that means for a listing

Treat a marketplace listing as advertising to people, which is what it is sold as, and treat the firm’s own listing page as the version of record. The page on the firm’s domain should hold everything the marketplace entry holds and more: the address written out, the size, the asking rate with its basis, the availability date, the building’s specification, the broker’s name and license, and the date the page last changed. If an assistant can read the marketplace as well, nothing is lost. If it cannot, the facts are still on a page it can.

How to test whether an assistant can open a listing

1. Paste the marketplace address of one of your listings into ChatGPT, Gemini, Copilot and Perplexity and ask for the size, the rate and the broker.
2. Do the same with the address of that listing on your own site.
3. Note which answers quote the page, which say the page could not be opened and which answer from somewhere else.
4. Repeat with a listing that closed last year, to see what lingers.
5. Keep the results with the date and rerun them each quarter, since platform defenses change without notice.

How do ChatGPT, Gemini, Copilot, Perplexity and Google’s AI features reach a commercial real estate website?

Each through an agent its maker documents, and each maker says what a refusal costs. A firm’s own site is the one place in this market where those rules are entirely its own decision.

Three gates: robots.txt, the firewall and the CDN

A robots.txt file that welcomes every crawler does nothing if a security product in front of the site turns them away, and the listing sites above show how common strict bot rules are in this industry. OpenAI and Perplexity both publish the IP ranges their crawlers use and recommend allowing requests from them. Anthropic publishes its source addresses too, and warns that blocking by IP address can stop its bots from reading robots.txt at all. Check all three layers for each agent by name. The documentation is short: OpenAI’s crawler page, Anthropic’s and Perplexity’s. The free AI crawler access checker on this site parses robots.txt for each of those agents.

Google’s own switch: the Search generative AI control

As of August 31, 2026, Google says, every Search Console property has a setting, under Settings and then Search generative AI, that decides whether the site’s links and content may appear in AI Overviews, AI Mode and the generative AI features in Discover. Include is the default. Exclude removes the site from those features and, by Google’s description, generally takes a few days to apply. Child properties follow their parent unless someone changes that. Google is explicit that the control is not a ranking signal elsewhere in Search and does not govern AI training, which is the job of Google-Extended. A marketing lead should know who in the company can reach that setting. Google’s help page for the control shows where it sits.

What Google says a site does not need

Google published a guide to generative AI search in May 2026, and part of it is a list of things site owners can ignore as far as Google Search is concerned: llms.txt files and other special machine-readable files, breaking content into small chunks, rewriting pages in a special way for AI, and chasing inauthentic mentions. It also says structured data is not required for its generative AI features, and that from Google Search’s perspective optimizing for generative AI search is still SEO. We agree with the substance. Nothing on this page is a trick aimed at a model. It is the ordinary discipline of publishing true, specific, dated facts on pages machines are allowed to read, applied to an industry that has not been in the habit. Google’s guide is worth reading in full before buying an AEO service from anyone, ourselves included.

Listing search and scripts

Many brokerage sites draw their listing search with JavaScript fed from a listing database. Google renders JavaScript, yet its own documentation still calls server-side or pre-rendering a great idea, partly because not all bots can run JavaScript. The cautious reading is that a fact which appears only after a script runs may not exist for some AI crawlers, so each listing, profile, availability table and report should have an address of its own that returns its facts as HTML text. The Ohio property search described above is a live example of a page whose records are not in its HTML.

What should a brokerage, landlord or developer publish to be named in AI answers?

Seven kinds of page, each built around facts someone outside the firm could check: broker profiles, submarket pages, availability tables, closed-transaction pages, property-type explainers with a glossary, research in HTML and a firm page. Most firms have a version of each. Few have them in a form a machine can use.

The search side of these pages, meaning which phrases they target and how they are linked, is covered in commercial real estate SEO, and the templates and listing database behind them in commercial real estate website design. The table below applies a different test to each one: what must be on it in words, and which outside record does it agree with?

Broker profiles that agree with the record

A profile is the page an assistant reaches for when it is asked about a person, and it is usually the thinnest page on a brokerage site. The content a profile needs is covered on the SEO page. The AEO addition is a habit of comparison: put each claim on the profile next to the record that confirms it, and fix whichever side is wrong.

• The name exactly as it appears on the license, with any everyday name beside it.
• License number and state for every state where the broker is licensed, typed as text, never placed inside an image or a badge.
• Designations written as words, such as CCIM designee or SIOR member, matching the directory entry.
• Property types and submarkets named the way clients say them, and the same way the press has described the broker.
• A link to the license record where the regulator gives it a stable address.
• The date the profile was last checked against those records.

In several states part of this is not optional. California’s statute, for one, requires a licensee’s name, license identification number and responsible broker’s identity on solicitation materials intended as a first point of contact, and it counts advertisements in electronic media among them. The website design page sets out four states’ disclosure rules side by side.

Submarket pages: dated figures and a named method

A market page earns a citation when its numbers can be repeated without embarrassment. That takes five things on the page, beside the figure and not in a footnote three screens away.

• The period: the quarter or month the figure describes, and the date the page was updated.
• The geography: what the submarket’s name covers, by boundary roads, ZIP codes or a map with a description in words.
• The inventory: which buildings are counted, for example multi-tenant office buildings above a stated size, with owner-occupied space in or out.
• The basis: the unit and the lease type the rate assumes.
• The method: whose data it is, how it was collected and how vacancy and absorption are defined.

The definitions do not have to be invented. The Commercial Real Estate Development Association’s glossary defines vacancy as physically vacant space over total existing inventory, with space under construction left out, and gross absorption as space occupied over a period before anything vacated is subtracted. Saying on the page that the firm follows those definitions, or where it departs from them, is what a named method means. The association’s terms and definitions are free to read and worth adopting as a house standard.

Availability tables: what is open now, and since when

“Who has 5,000 square feet available in this submarket” is a question about today, and the commonest wrong answer is a suite that leased months ago. An availability table on the building’s page fixes that if it is kept honestly. Give each suite a row with its size, its asking rate and basis and the date it is available, and print the date the table was last checked above it. Mark sublet space as sublet, since the glossary defines it as space offered by a tenant and not by the landlord. Remove a suite the day its lease is signed, and keep the building page itself, because availability changes and the building does not.

Closed-transaction pages

A closed deal is the strongest evidence a brokerage has, and most firms bury it in a news release or a tombstone graphic. Give each significant transaction a page that states, in sentences, the address, the property type, the size, whether it was a sale or a lease, the month it closed, which side the firm represented and who at the firm worked on it. Confirm with the client what may be disclosed; price and tenant names often may not be. Then keep the page honest over time. Texas, for one, treats an advertisement about a property made ten days or more after closing as misleading unless it gives the property’s current status, and treats the same way any advertisement implying that a person was involved in a transaction in which they had no role. Label closed deals as closed, with the date.

Explainers and a glossary that can be quoted one entry at a time

Assistants are asked what flex space is, how a power center differs from a strip center and what a load factor does to a quoted rent. One page per property type is enough; Google’s guide warns against creating separate content for every variation of a search, so resist publishing a near-copy for every city. A glossary works for AEO when each entry can be lifted out whole: the term, then one sentence that defines it without leaning on the entries around it, then an example with numbers. Put each term under its own heading with its own anchor, so a citation can point straight at it, and say where the local market uses a term differently. That local difference is exactly what a national glossary cannot supply.

Research without the form in front of it

A report released only after an email address is entered is not on the open web, whatever its quality. Publish the findings, the tables and the method as a page and attach the designed PDF as a download; the form can stay on the download. The SEO page covers how a report series is structured to rank. Content marketing covers how research is produced.

What changes for owners, landlords and developers?

The method stays the same; the pages differ. A brokerage is asked about people and markets. An owner is asked about buildings. A developer is asked about projects that do not exist yet.

Owners and landlords: the building as the subject

A building is the thing a tenant asks an assistant about, by name or by address. Give each one a permanent page that states the address the way the county assessor writes it, the year it was built, the size, the floor plates, the specification that matters for its property type, what is available today and whom to call. Keep the page when the building is full. Where a leasing brokerage markets the property, agree which page is the complete one, and make the two say the same thing. Where the building is also in an economic development database, make that entry say it too.

Developers: projects that are discussed before they open

A development is talked about by people who are not its customers: neighbors, local officials, lenders, journalists, future tenants. Any of them may ask an assistant what is being built. A project page should answer in text: the location, the uses, the size in square feet or units, the stage the project has reached with the date of that statement, the development team and a contact. Use stage words in the sense the industry uses them; the glossary counts a building as under construction once permits have been obtained and site excavation has begun, and notes that demolition alone does not mean construction has started. Update the stage when it changes and leave the earlier dates visible. A rendering with no words beside it tells a model nothing.

• Location: the street address or the bounding streets, the municipality and the county.
• Use: each component by type, with its size.
• Stage: planned, approved, under construction or delivered, each with the date the statement was made.
• Team: developer, architect, general contractor and leasing contact, named with their permission.
• Documents: public filings and approvals described in a sentence each, in place of a bare download link.
• What is not yet decided, said plainly, so that nobody fills the gap with a guess.

Public REITs and private owners

A REIT registered with the SEC has a second public record in its filings, and the two should never disagree about what the company owns. A private owner has no filings to be measured against, which leaves its own pages, the county’s land records and the announcements of its tenants and lenders as the whole of what can be read. For portfolios large enough to need sign-off from several departments, enterprise AEO covers how the governance is usually arranged.

How should a market figure be written so an assistant can repeat it safely?

With its date, its geography, its basis and its owner inside one sentence. An assistant shows a sentence or two, not the chart, the footnote or the paragraph that came before. Whatever the sentence leaves out is lost.

The table shows the difference. Firm names and figures are placeholders in square brackets, since no real figure belongs on a page like this one.

Four parts every figure needs

• When: the period measured and, separately, the date published.
• Where: the named area, defined once on the page.
• What: the unit, the lease basis and the buildings counted.
• Whose: the firm or outside body that produced the number.

Why two firms publish two vacancy rates for one submarket

Both can be right. The industry glossary notes that data providers each have their own set of buildings in the competitive inventory, and that owner-occupied and government-owned buildings are typically excluded. A firm that counts multi-tenant buildings above 20,000 square feet and a firm that counts everything above 10,000 will publish different rates for the same streets. An assistant that meets both figures without the inventory stated beside them can only pick one, or average them, which is worse. Stating the inventory in the sentence is what keeps the firm’s number from being called wrong.

Name the firm inside the sentence

“Our research shows” loses its owner as soon as it is quoted, because the reader of the answer cannot tell whose research it is. Writing the firm’s name into the statement feels stiff on the firm’s own page. It is also the change that lets the sentence travel with its owner attached.

Name the area the way the rules expect

Submarket names are marketing as well as geography, and an assistant repeats them without knowing which. New York’s advertising regulation requires an advertisement that places a property in the vicinity of an area to name the area where the property actually is, and prohibits names that would mislead the public about location. Wherever a listing or market page borrows a better-known neighborhood, put the true municipality or submarket in the same sentence.

A checklist before a market page goes live

1. Does every figure have its period beside it?
2. Is the submarket defined in words somewhere on the page?
3. Is the inventory stated: which buildings, above what size?
4. Is the unit stated each time: per year or per month, gross or net?
5. Is the method described in a short paragraph of its own?
6. Are the tables real HTML tables, not pictures of tables?
7. Does any sentence about the future read as a promise?
8. Is the firm named in the sentences most likely to be quoted?
9. Is last quarter’s page still available at its own address, labeled with its period?

AEO content writing applies the same pattern to whole pages, and our free extractable content checker shows whether a page’s answers survive being pulled out on their own.

Which structured data helps an assistant tie a commercial page to a public record?

A little, and none of it is a requirement. Google states that structured data is not required for its generative AI features and that there is no special schema.org markup to add. Its value on a commercial site is consistency: the markup, the visible text and the outside record all saying one thing. The per-template markup for search, from Organization to RealEstateListing, is set out on the commercial real estate SEO page.

Credentials on a broker’s profile

Schema.org gives a person or an organization a hasCredential property, and its EducationalOccupationalCredential type covers a diploma, degree, certification, qualification or badge awarded to someone who meets the credentialer’s requirements. A license or a designation fits that description. No Google search feature is documented for it; the point is that the markup, the visible text and the regulator’s record all say one thing. The hasCredential definition is one paragraph long.

What schema.org does not have

There is no schema.org type for an office building or a warehouse. ShoppingCenter and SelfStorage exist, as kinds of local business. The properties that carry floor size, year built and permitted usage sit under Accommodation, which schema.org defines as a place that can accommodate human beings, with a hotel room and a meeting room among its examples. For most commercial buildings the honest choice is Place, with the address and coordinates, and the specification written in the page’s text.

Markup has to match the page

Google’s structured data guidelines say not to mark up content that is not visible to readers, and not to mark up misleading content. If the markup carries a license number, the number is printed on the page. If it carries a designation, the designation is in the text. Test templates with Google’s Rich Results Test and the Schema Markup Validator, and use our free schema and copy validator to compare markup with visible text.

FAQ markup after May 2026

Google’s documentation updates record that the FAQ rich result stopped appearing in Search on May 7, 2026, and that the feature’s documentation was removed the following month. FAQPage remains a valid schema.org type, and a question with its answer directly beneath it remains a useful way to write, which is why this page has both. Nobody should now be sold FAQ markup as a route to a larger search listing.

Which rules limit what a commercial real estate page can claim?

State brokerage advertising rules, the FTC’s rule on reviews and testimonials, the Fair Housing Act wherever housing is advertised, and securities law wherever an investment is offered. A sentence written to be quoted by an assistant is still an advertisement, and the most quotable sentences, the rankings, the records and the returns, are the ones these rules reach first. What follows is general information for marketing teams, not legal advice.

State advertising rules vary, so read your own commission’s

Brokerage is regulated state by state, and each commission writes its own advertising rule. The specifics in the table come from California, New York and Texas and should not be assumed anywhere else. What they share is the point that matters for AEO: none of them cares where a sentence is read. A ranking in a profile, a value in a listing or a borrowed place name is just as much the firm’s advertisement when an assistant repeats it to a stranger. The texts are short: California section 10140.6, New York’s section 175.25 and the Texas Real Estate Commission’s rules. The disclosure lines each state asks for are tabulated on the website design page and the advertising rules on the marketing agency page. A firm working in other states should read its own commission’s rule before borrowing any of this.

Reviews: the FTC’s rule on fake reviews and testimonials

Assistants asked to compare firms draw on what clients say about them, which makes reviews part of AEO whether a firm likes it or not. The FTC’s Rule on the Use of Consumer Reviews and Testimonials took effect on October 21, 2024, and the Commission notes that it lets courts impose civil penalties for knowing violations. It prohibits a business from writing, creating or selling a review that misrepresents whether the reviewer exists or had experience with the business; from offering anything in exchange for reviews that express a particular sentiment; from publishing reviews by its own officers or managers without a clear disclosure of the relationship; from presenting a review site it controls as independent; and from suppressing reviews through groundless legal threats, intimidation or false public accusations. Google added a guideline on fake and undisclosed incentivized reviews to its review snippet documentation in July 2026. For a brokerage or landlord the safe practice is plain: ask every client the same way after a deal, offer nothing that depends on the rating, keep staff and their families out of the review pages, and answer criticism without threats. The FTC’s questions and answers and the rule text are written for businesses.

The Fair Housing Act, where apartments are advertised

Offices and warehouses are not dwellings, but apartments are, and many commercial firms sell, lease or manage them. The Act defines a dwelling as a building, or part of one, occupied or intended as a residence, and makes it unlawful to publish any notice, statement or advertisement about 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 regulation adds that this reaches words, phrases, photographs, illustrations and symbols, and the choice of media or locations for advertising. For AEO the risk is the summary: an assistant describing a multifamily listing repeats its adjectives, so write about the building, the units, the costs and the facilities nearby, and never about who lives there or who would suit it. Section 3604 and 24 CFR 100.75 are the texts, and the marketing agency page works through how the regulation applies to campaigns.

Offerings and track records

A sponsor raising capital for a property chooses, with counsel, which exemption the offering relies on, and that choice decides what may be said in public. Under Rule 506(b) the company cannot use general solicitation or advertising to market the securities; under Rule 506(c) it can, provided every investor is accredited and the company takes reasonable steps to verify that. Regulation D’s own examples of general solicitation are advertisements and articles in newspapers, magazines and broadcast media, and seminars whose attendees were invited by general advertising, and the rule says the list is not exhaustive. A page open to anyone can be quoted by an assistant to anyone. The SEC’s investor site sets out both versions of Rule 506; the marketing agency and lead generation pages cover what each allows in a campaign.

Return figures are the most quotable sentences a sponsor has, and the most regulated. Investment advisers registered or required to be registered with the SEC are bound by the Marketing Rule, which prohibits advertisements that include an untrue statement of a material fact and permits gross performance only when net performance is shown with at least equal prominence, calculated over the same period. State the period, state whether the figure is gross or net, state what it covers, and publish nothing a model could repeat as a promise. The Marketing Rule’s text lists the conditions.

Commercial real estate AI: software tools, or being named in answers?

Mostly software. Most of the interest in AI and commercial property comes from people looking for commercial real estate AI tools and the vendors that sell them, rather than from firms that want to be found.

Two meanings of AI in commercial real estate

A broker who goes looking for AI help usually wants a tool from one of the commercial real estate AI companies: something to draft an offering memorandum, abstract a lease or build a model. That is AI for commercial real estate brokers as an instrument. This page is about the other direction, AI as the place where the broker’s next client asks a question. The two meet at one point. A firm that uses assistants in its own work has every reason to assume that its clients, lenders and counterparties do too.

Taken together, this does not argue against the page. It does say the service is early; most interest in AI in commercial real estate is still about software. Few brokerage owners ask for it by name yet, while every question in the tables above can be put to an assistant today and will get an answer that names somebody.

How is retrievability checked on a commercial real estate site?

Compare what a crawler actually receives with what the firm thinks it published. Every check below can be run without special access, and most are quick.

1. Fetch the site’s robots.txt and read it for OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot and Bingbot by name, and for any blanket rule that catches them.
2. Ask whoever runs the firewall or CDN whether bot rules are switched on, and whether the published crawler IP ranges are allowed.
3. In Search Console, open Settings and confirm that the Search generative AI control is set to include.
4. Request one broker profile, one listing, one availability table and one market page from the command line and read the HTML: are the license number, the rate and the vacancy figure present as text?
5. Turn JavaScript off in a browser and load the listing search. Note what disappears.
6. Check the templates for a stray noindex, nosnippet or noarchive tag.
7. Search Google for a distinctive sentence from a market page, in quotation marks, to confirm the page is indexed.
8. Open the latest research report without filling in a form. If that cannot be done, the report is not public.
9. Look up three brokers in their state license lookups and compare name, number and status with their profiles.
10. Look up each designee in Find a CCIM and in the SIOR directory.
11. Find each building or site the firm markets in the state or regional economic development database, and compare size, acreage and contact with the firm’s own page.
12. Search the trade press for the firm’s last five announced deals and compare sizes, dates and roles with the deal pages.
13. Compare the firm’s name, addresses and phone numbers across the website, Google Business Profile, Bing Places and the listing marketplaces.
14. Ask each assistant the full prompt set and file every answer under its date.

Our AI visibility audit runs this whole list and writes it up; technical AEO takes on crawler and rendering fixes; the free Google Business Profile and NAP audit does the name, address and phone comparison in step thirteen.

Reading a page the way a crawler does

Loading a profile in a browser shows the finished page, scripts and all. Fetching the same address from a terminal with curl returns the raw HTML a crawler begins with. Look for five things in that response: the title, the meta description, the main heading, the body text containing the fact the page exists to state, and any robots meta tag. If the license number or the rent figure is not in that text, treat it as unpublished.

How are results measured?

Monthly, by asking the same questions in the same words, and by reading the two first-party reports that reached site owners in 2026: Google’s generative AI performance report in Search Console and Bing Webmaster Tools’ AI Performance report. No ranking position exists to report, since each answer is composed fresh.

What a prompt set for a commercial firm contains

• Recommendation prompts by property type and submarket: who leases industrial space in a named corridor.
• Price prompts: what a given size of space costs in each submarket the firm covers.
• Availability prompts: who has space of a given size open now in a named submarket.
• Market prompts: vacancy, absorption and cap rates for the firm’s property types.
• Site prompts: sites or buildings of a given size with given utilities in the counties the firm serves.
• Definition prompts: the lease terms the glossary covers.
• Verification prompts: whether a named broker is licensed, whether a named firm exists, what a named sponsor has filed.
• Building prompts: what is known about properties the firm has listed or sold.

A first draft of the prompt set can come from our free AI visibility prompt builder, and the AI visibility checker gives a quick first reading. AI visibility tools compares the paid trackers, and why ChatGPT recommends a competitor lists the usual causes when another firm is named instead.

Reading the two first-party reports

Google’s generative AI performance report counts impressions: the times a link to the site was shown inside AI Overviews or AI Mode. It can be viewed by page, which tells a brokerage whether its market pages, its profiles or its listings are the ones being surfaced. Microsoft’s AI Performance report counts citations in Copilot and Bing’s AI-generated summaries and shows a sample of the grounding queries behind them, which is the nearest thing either company publishes to the wording assistants use. Google’s guide adds a warning worth repeating to anyone selling dashboards: no third-party tool has access to Google’s internal ranking or AI systems. Google’s help page for the report and Microsoft’s announcement describe what each counts.

What should a firm do when an assistant states something wrong about it?

Fix the fact where the assistant read it, then wait for the recrawl. Wrong answers about commercial firms usually trace back to a page: a profile that still lists a broker who left, a marketplace entry for a building that sold, an old news release, a database entry with last year’s size, a directory with a previous address.

1. Save the answer word for word, with the prompt, the assistant and the date.
2. Open the links it shows, if it shows any, and find the sentence the error came from.
3. Correct the firm’s own page first, and add the date of the correction.
4. Correct the outside record: the directory entry, the marketplace listing, the economic development database, the business profile.
5. Where the license record itself is out of date, update it with the regulator. That is a licensing matter before it is a marketing one.
6. Ask again in the next monthly run and record whether the answer changed.

Two limits are worth stating. An assistant that answers from memory without showing links leaves no trail to follow, and nobody can promise when a corrected page will be read again. The firm controls its own pages and the record of its own diligence, and that is where the effort goes. Online reputation management covers the cases where the wrong information sits on someone else’s site.

What does AEO work for a commercial real estate firm include?

Four strands: a survey of what machines can reach now, rebuilt pages carrying checkable facts, a compliance review of every claim, and monthly measurement. All four are set out in a written scope before anything starts.

1. Baseline: every prompt asked in every assistant once, and each answer kept verbatim.
2. Access audit: robots.txt, firewall, CDN, the Search Console control and how listing and profile pages render without scripts.
3. Source review: the firm’s entries on marketplaces, economic development databases and directories, and its recent press, compared with its own pages.
4. Register check: each broker’s license record and designation listing compared with the site, and entity names compared with filings.
5. Profile rebuild: license numbers, designations, specialties and transactions as text on every broker page.
6. Market and availability pages: dated figures with period, geography, inventory and method, and availability tables with a checked date.
7. Transaction pages: closed deals written up with the client’s approval, labeled with status and date.
8. Explainers, glossary and research: property-type pages, lease terms and the latest report published as HTML.
9. Compliance pass: state advertising rules, review practice, fair housing wording on any housing page and counsel’s review of anything that touches an offering.
10. Every month: the prompt set again, both first-party reports, the error log and the corrections it calls for.

How long before anything changes?

A corrected profile or a new market page is live the day it is published. What follows is in nobody’s gift: search engines recrawl on their own schedules and assistants refresh on theirs. Perplexity says a robots.txt change can take up to 24 hours to register with its systems, and Google says a change to its generative AI control generally takes a few days. No one can date a particular answer beyond that. What can be promised is the work itself and the monthly reading of results; how long AEO takes covers the timing in more depth.

AEO, GEO or AI SEO for commercial real estate: what is the work called?

Several names cover one job. The vocabulary multiplied faster than the work did, and Google’s position is that all of it remains SEO. The scope on this page is the same whichever term a search started from.

One caution on the abbreviations: in this industry CRE already means commercial real estate and GEO is easily read as geography, so AEO is the least confusing label of the three. AEO vs GEO vs LLM SEO compares the definitions, and the generative engine optimization, LLM SEO and AI SEO pages approach the work from each label.

What does AEO for commercial real estate cost?

As published price bands: $1,000 to $4,000 for a one-off AEO audit, and $1,500 to $20,000 a month for a retainer, with the number of offices, brokers and submarkets deciding where in that span a firm lands. Every quote is built from a written scope.

The bands are the ones in our AEO pricing guide and marketing agency pricing guide, published for planning rather than as quotes. Inside a band, the price follows the count of broker profiles to rebuild, submarkets that need a data page and buildings that need availability tables, plus whether research already exists and how much of the listing system must change. The one-off option is detailed on the AEO audit page, and smaller scopes on affordable AEO services.

How do you choose a provider for commercial real estate AEO?

Ask what they would check before what they would write. A provider that starts with content volume has not understood an industry where the scarce thing is a fact someone else can confirm.

Questions to put to any provider before signing

• Which of our pages would you rebuild first, and what would be on them that is not there now?
• What does one of our broker profiles return when it is requested as a crawler would request it?
• Which of our published figures could not be repeated safely as written?
• Where does our largest vacant building appear outside our own site, and can an assistant read those places?
• What would you refuse to publish for a sponsor with an open offering?
• If no answer moved this month, what would your report still tell us?
• If we part ways, what do we keep, and in what format?

When a firm should spend the money elsewhere

A brokerage that prefers to keep its deals, its people and its market views out of public view leaves an assistant nothing to cite, and should not buy this. The same goes for a sponsor whose counsel has ruled out public marketing, and for a firm without a website of its own. For them the better spend is accurate license and directory records, a complete Google Business Profile for each office and local SEO. How to choose an AEO agency and our AEO consultant page go through the broader choice.

Where does this sit next to commercial real estate SEO, lead generation and the website?

On the same pages. A market page with dated figures ranks, gets quoted, gives a paid campaign somewhere honest to land and gives a broker something to send. Each program is measured differently; the writing they depend on is the same.

Commercial real estate SEO covers ranking for markets, property types and names. Commercial real estate lead generation covers how inquiries are produced and qualified. Commercial real estate website design covers the listing database, the templates and the confidentiality gate for offering materials. Commercial real estate marketing is the whole program. LinkedIn advertising, email marketing and outbound lead generation carry the research and the deal announcements to named accounts, and AEO vs SEO marks where the two disciplines diverge.

Neighboring markets with pages of their own

• AEO for real estate agents: residential agents, teams and brokerages.
• AEO for property management: fees, licensing and the questions owners ask.
• AEO for apartments: communities, floor plans and leasing questions.
• AEO for construction and AEO for architects: the firms that design and build the projects.
• AEO for home builders: residential development.
• AEO for mortgage brokers and AEO for financial services: debt and capital.
• AEO for law firms: the counsel across the table in a lease or a sale.
• AEO for hotels and AEO for senior living: operating property types.
• AEO for B2B and AEO for industrial companies: the occupiers who lease the space.

Related services

Where a commercial real estate firm usually goes from here, beginning with the SEO page this one pairs with.

• Commercial real estate SEO: how the same market, listing and profile pages are built to rank in Google.
• Answer engine optimization agency: how Progression runs AEO across every industry.
• Commercial real estate marketing agency: the whole commercial program, from marketplaces to offering packages.
• Commercial real estate lead generation: turning tenant, owner and investor interest into qualified inquiries.
• Commercial real estate website design: listing databases, broker pages and the memorandum gate.
• Real estate marketing, real estate SEO and real estate PPC: for residential agents and brokerages.
• Property management marketing and property management SEO: for third-party managers.
• Apartment marketing: filling units in a lease-up.
• Real estate website design and real estate software development: agent sites and custom property software.
• B2B marketing and B2B PR: long, committee-driven sales and trade coverage.
• Local AEO: answers to questions that name a town, county or submarket.
• How to get cited by AI and how AI search works: what retrieval and citation actually involve.
• What is AEO and AEO myths: a primer, and the claims worth doubting.
• AEO services and enterprise AEO: engagements from a single office up to a national platform.
• LLM visibility and how to rank in ChatGPT: the topic seen from the model’s end.
• AEO tools and AI visibility trackers compared: prices and coverage of the tracking software.
• FAQ schema generator and llms.txt generator: two free utilities; read the cautions above before relying on either.
• Why AI visibility dropped and AI answers vs organic search: diagnosing a fall in mentions or clicks.
• Public relations and lead generation: coverage earned from others, and the follow-up once someone calls.
• SEO content writing and technical SEO: copy and the site machinery under it.
• Construction marketing, architecture firm marketing and marketing for home builders: for the firms that design and build.
• Hotel marketing and senior living marketing: operating property types.
• Landing page design and B2B website design: pages made for a cautious, considered buyer.
• International AEO: for firms with offices or capital abroad.
• AEO by industry: the full list of market-specific AEO pages.

Frequently asked questions