A practical ASEO and GEO guide for estate agents, letting agencies, and property portals. JSON-LD schema templates, CPS® content audit framework, and AI citation optimisation for residential sales, lettings, and commercial property businesses.
When a buyer asks ChatGPT "best estate agents in Gloucester" or Perplexity "letting agents near me for a two-bed flat", AI systems construct answers from structured, citable sources. Most estate agent websites fail to appear in these answers for three reasons: missing or incorrect JSON-LD schema, unstructured content that AI cannot parse, and no location enrichment data to answer proximity or neighbourhood queries.
An estate agent with a well-optimised website can rank on page one of Google and score an F on the Citation Probability Score® - meaning it is completely invisible to AI search despite strong traditional SEO performance. These are different retrieval mechanisms requiring different optimisation strategies.
| Business type | Correct schema type |
|---|---|
| Estate agent or letting agency | RealEstateAgent |
| Individual property listing | RealEstateListing |
| Open house or viewing event | Event |
| Neighbourhood or area guide | Place or AdministrativeArea |
Using LocalBusiness for an estate agent is incorrect. AI systems filter
by schema type when answering discovery queries. A business using LocalBusiness
will not appear for "estate agent" or "letting agent" queries in AI-generated answers.
{
"@context": "https://schema.org",
"@type": "RealEstateAgent",
"name": "Your Agency Name",
"description": "Independent estate and letting agency in [city], serving [area list]. Specialising in residential sales, lettings, and property management across [key neighbourhoods].",
"url": "https://yourwebsite.com",
"telephone": "+44-000-000-0000",
"address": {
"@type": "PostalAddress",
"streetAddress": "123 High Street",
"addressLocality": "Your Town",
"addressRegion": "Your County",
"postalCode": "AA1 1AA",
"addressCountry": "GB"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": 00.0000,
"longitude": -0.0000
},
"areaServed": [
{
"@type": "City",
"name": "Your City"
},
{
"@type": "AdministrativeArea",
"name": "Neighbourhood 1"
},
{
"@type": "AdministrativeArea",
"name": "Neighbourhood 2"
},
{
"@type": "AdministrativeArea",
"name": "Nearby Town"
}
],
"knowsAbout": [
"residential sales",
"lettings",
"property management",
"first-time buyers",
"buy-to-let",
"leasehold properties"
],
"aggregateRating": {
"@type": "AggregateRating",
"ratingValue": "4.8",
"reviewCount": "124"
},
"potentialAction": {
"@type": "ReserveAction",
"name": "Book a valuation",
"target": "https://yourwebsite.com/valuation"
}
}Individual property listing pages should use RealEstateListing schema.
This makes listings citable for specific property queries such as
"two-bed flat to rent in [area] under £1,200 per month."
{
"@context": "https://schema.org",
"@type": "RealEstateListing",
"name": "Two-Bedroom Flat in [Neighbourhood], [City]",
"description": "Two-bedroom ground floor flat in [Neighbourhood], [City]. 650 square feet. 5-minute walk to [train station], 3 minutes to [supermarket]. EPC rating B. Available from [date]. Pets considered.",
"url": "https://yourwebsite.com/property/listing-id",
"datePosted": "2026-03-01",
"about": {
"@type": "Apartment",
"numberOfBedrooms": 2,
"floorSize": {
"@type": "QuantitativeValue",
"value": 65,
"unitCode": "MTK"
},
"address": {
"@type": "PostalAddress",
"streetAddress": "Flat 2, 45 Example Road",
"addressLocality": "Your Town",
"postalCode": "AA1 1AA",
"addressCountry": "GB"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": 00.0000,
"longitude": -0.0000
}
},
"offers": {
"@type": "Offer",
"price": 1150,
"priceCurrency": "GBP",
"priceSpecification": {
"@type": "UnitPriceSpecification",
"price": 1150,
"priceCurrency": "GBP",
"referenceQuantity": {
"@type": "QuantitativeValue",
"value": 1,
"unitCode": "MON"
}
}
}
}The areaServed field is the most impactful schema addition for estate agents.
AI systems answering "estate agents covering [village]" or "letting agents
in [neighbourhood]" filter by areaServed. An agency that only lists its
office town will be invisible for all surrounding area queries. List every
town, village, and neighbourhood the agency genuinely covers.
The knowsAbout field signals specialist expertise to AI systems. An agency
that lists "first-time buyers", "buy-to-let", and "leasehold properties" in
knowsAbout will appear for queries like "estate agents who specialise in
first-time buyers in [city]" — queries a generic listing will never surface for.
AI systems answering proximity queries need enrichment data beyond a street address. The difference between a listing AI ignores and one it cites is often the presence of specific, structured proximity data.
Add the following as additionalProperty fields in schema and write the
same information into the visible page content:
{
"@type": "Apartment",
"additionalProperty": [
{
"@type": "PropertyValue",
"name": "Nearest train station",
"value": "[Station Name] — 5 minutes walk"
},
{
"@type": "PropertyValue",
"name": "Nearest supermarket",
"value": "[Supermarket Name] — 3 minutes walk"
},
{
"@type": "PropertyValue",
"name": "Schools within 1km",
"value": "2 primary schools, 1 secondary school"
},
{
"@type": "PropertyValue",
"name": "City centre commute",
"value": "12 minutes by train, 20 minutes by bus"
},
{
"@type": "PropertyValue",
"name": "Walkability",
"value": "High — daily errands achievable on foot"
}
]
}The homepage should answer three AI queries in its opening 167 words: what the agency does, where it operates, and what makes it different.
Before (fails CPS® Answer Structure pillar): "Welcome to [Agency Name]. We are a passionate team of property professionals dedicated to helping you find your perfect home or achieve the best possible price for your property..."
After (passes CPS® Answer Structure pillar): "[Agency Name] is an independent estate and letting agency based in [Town], covering [area list]. The agency handles residential sales, lettings, and property management, with an average sale achieved at 98.6% of asking price and an average time to sale of 34 days. [Agency Name] has managed over 400 property transactions since 2018 and holds a 4.8-star rating across 124 verified reviews."
The second version gives AI systems six facts to extract and cite. The first contains no citable information.
Each listing page description should be structured as follows across 134 to 167 words:
- First sentence: property type, bedrooms, neighbourhood, city, price
- Second sentence: size, key features, aspect or floor
- Third sentence: proximity to transport, schools, and amenities
- Fourth sentence: availability, EPC rating, and next step
This structure ensures AI can answer "find a [property type] in [area] under [price]" by extracting the first sentence, while the full block answers more detailed queries.
Area guides are the highest-value content type for estate agent AI citation. AI systems answering "what is [neighbourhood] like?" or "best areas to live in [city] for families" draw heavily from area guide content. Each area guide should cover:
- Average property prices (sale and rental) with a date
- Transport links with specific journey times
- Schools with Ofsted ratings where available
- Parks, green spaces, and recreation
- Shops, restaurants, and daily amenities
- Demographic character of the area
- Recent development or regeneration activity
Each section should be 134 to 167 words and open with a direct answer to the most common question about that topic.
Every page should include five questions matching what buyers and renters actually ask AI systems. Examples:
- "How long does it take to sell a house with [Agency Name]?"
- "Does [Agency Name] manage properties as well as sell them?"
- "What areas does [Agency Name] cover?"
- "What fees does [Agency Name] charge for lettings?"
- "Is [neighbourhood] a good area to buy in [city]?"
AI systems are increasingly answering "open houses this weekend in
[area]" queries. Use Event schema for every viewing event:
{
"@context": "https://schema.org",
"@type": "Event",
"name": "Open House: Two-Bedroom Flat in [Neighbourhood]",
"startDate": "2026-04-05T10:00:00+01:00",
"endDate": "2026-04-05T12:00:00+01:00",
"eventAttendanceMode": "https://schema.org/OfflineEventAttendanceMode",
"location": {
"@type": "Place",
"name": "45 Example Road, [Town]",
"address": {
"@type": "PostalAddress",
"streetAddress": "45 Example Road",
"addressLocality": "Your Town",
"postalCode": "AA1 1AA",
"addressCountry": "GB"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": 00.0000,
"longitude": -0.0000
}
},
"organizer": {
"@type": "RealEstateAgent",
"name": "Your Agency Name",
"url": "https://yourwebsite.com"
},
"description": "Open house viewing for a two-bedroom flat in [Neighbourhood]. No appointment needed. Viewing runs from 10am to 12pm. Contact [agent name] on [number] for more information.",
"offers": {
"@type": "Offer",
"price": 0,
"priceCurrency": "GBP",
"availability": "https://schema.org/InStock"
}
}1. Using LocalBusiness instead of RealEstateAgent schema
AI systems filter by schema type. An estate agent using LocalBusiness
will not appear for agent discovery queries regardless of content quality.
2. No areaServed field
Without areaServed, an agency is invisible for any query mentioning
a town or neighbourhood it covers but has not listed in schema.
3. Listing descriptions under 100 words Short descriptions give AI systems nothing to parse. Every listing page needs a minimum of 134 words of structured prose — not bullet points.
4. No area guide pages Area guides are the single highest-leverage content type for estate agent AI citation. Agencies without them miss every neighbourhood and lifestyle query AI receives.
5. No FAQ sections FAQ content is cited at a significantly higher rate than standard prose because the format matches how AI systems construct responses. Every page should include at least five relevant questions with direct answers.
What is GEO for estate agents?
GEO (Generative Engine Optimisation) for estate agents is the practice of
structuring website content and schema markup so that AI systems — including
ChatGPT, Perplexity, and Google AI Overviews — can accurately retrieve and
cite the agency when buyers or renters ask AI for property or agent
recommendations. For property businesses it focuses on three areas: correct
schema type (RealEstateAgent), location enrichment for proximity queries,
and CPS®-structured content on listing and area guide pages.
How is property GEO different from traditional estate agent SEO?
Traditional estate agent SEO optimises for Google rankings using keyword density, backlinks, and Rightmove or Zoopla portal listings. Property GEO optimises for AI citation using schema markup, structured listing descriptions, and area guide content. An agency can dominate Rightmove search and be completely absent from ChatGPT and Perplexity answers for the same area. Both strategies require separate, parallel approaches.
Which pages should an estate agent optimise for AI citation first?
In order of impact: (1) the agency homepage, (2) area guide pages, (3) individual property listing pages, (4) agent profile pages. The homepage and area guides generate the highest volume of AI citations because they answer the discovery and research queries AI receives most frequently.
The full CPS® audit for estate agents and property businesses covers:
- Per-block CPS® scores across homepage, listing pages, and area guides
- Schema audit identifying incorrect types, missing fields, and markup errors
- Location enrichment gap analysis for all listed properties
- Share of Voice measurement across ChatGPT, Perplexity, Google AI, Gemini, Copilot
- Hallucination detection — identifying where AI misrepresents the agency
- Competitor citation analysis — which local agencies are being cited instead
Book a full audit: citedbyai.info
CPS® (Citation Probability Score®) and Cited By AI® are registered trademarks of Cited By AI, United Kingdom.