of UK property firms are invisible in AI search answers.
We asked ChatGPT the 128 questions that real UK property buyers search. We ran every question three times, across six buyer personas, with live web search forced on every answer. These are the brands it named. And didn't.
One image. Six buyer personas. The whole story.
Every row is a UK property firm. Every column is a type of buyer we primed ChatGPT with. The darker the cell, the more often the firm appears when that buyer asks the AI for advice. Look how sharply commercial brands stop at the residential boundary — and how most rows have white cells everywhere but one.
| Firm | Downsizer | First-Time | BTL | Intl. | Relocator | Commercial |
|---|---|---|---|---|---|---|
| Universal tier · appears for every persona | ||||||
| 01Rightmove | ||||||
| 02Zoopla | ||||||
| 03OnTheMarket | ||||||
| 04Savills | ||||||
| 05Knight Frank | ||||||
| 06Foxtons | ||||||
| 07Purplebricks | ||||||
| 08Connells Group | ||||||
| 09Hamptons | ||||||
| 10Yopa | ||||||
| 11PrimeLocation | ||||||
| Persona-specific tier · conditional on who's asking | ||||||
| 12Strutt & Parker | ||||||
| 13Chestertons | ||||||
| 14Dexters | ||||||
| 15Barnard Marcus | ||||||
| 16Beauchamp Estates | ||||||
| 17CBRE | ||||||
| 18JLL | ||||||
| 19Cushman & Wakefield | ||||||
| 20John D Wood & Co | ||||||
| 21OpenRent | ||||||
| 22Russell Simpson | ||||||
| 23Winkworth | ||||||
| 24Countrywide | ||||||
| 25William H Brown | ||||||
| 26Your Move | ||||||
| 27Reeds Rains | ||||||
| 28McCarthy Stone | ||||||
| 29Sequence | ||||||
| 30KFH | ||||||
| 31Colliers | ||||||
| 32CoStar | ||||||
| 33Rokstone | ||||||
| 34EG Propertylink | ||||||
| 35Halifax | ||||||
| 36WeWork | ||||||
| 37Regus | ||||||
| 38IWG | ||||||
| 39Coutts | ||||||
| 40EIG | ||||||
Top 40 firms by total AI mentions across all personas. Cell value = share of that persona's runs mentioning the firm, rounded to the nearest percent. Hover any cell for exact share.
For two whole categories of the UK property industry, ChatGPT has nothing to say.
When a UK buyer asks ChatGPT for a firm, four out of five firms in the industry don't get named — ever. Property management and chartered surveyors collectively field 53 of the UK's mid-market brands. ChatGPT surfaces zero of them.
The zero-visibility categories
Landscape of 298 firms built from 38 category SERP seeds, DR- and traffic-filtered, then Claude-classified. Visibility measured against 384 AI answers.
The same question asked six different ways returns six different rankings.
We primed ChatGPT with the demographic, financial and concern profile of each buyer before asking the same property questions. The top 10 it returns shifts substantially with the persona. Pick your reader.
First-Time Buyer
BTL Investor
International Buyer
The Relocator
Commercial Tenant
Commercial and residential property are two separate universes in AI answers.
CBRE, JLL, Cushman & Wakefield and Colliers dominate commercial tenant answers but disappear entirely from residential ones. Only Savills and Knight Frank bridge both sectors. Every other firm we observed lives in one universe, never the other.
Cross-sector firms identified as any firm appearing in 3+ runs for both a residential persona and the Commercial Tenant persona. Only Savills and Knight Frank meet the bar.
Scale buys AI visibility. It caps at 51%.
Even for national chains with hundreds of branches and decades of brand equity, ChatGPT only cites them about half the time. For regional and boutique firms, most are invisible. For micro firms, almost none surface at all.
Tier definitions — National: multi-region UK operator. Regional: multi-office inside a single UK region. Boutique: 1–3 offices with premium positioning. Micro: single office or founder-led. 9 firms unclassified; excluded from the chart. Source: landscape v2.
Eleven brands appear for every buyer. Everyone else fights for a fraction.
These 11 are the only firms that surface in AI answers across all 6 buyer personas we tested. Four are portals. Two are online/hybrid. Two are national networks. Three are household-name estate agents. If you're not here, AI search has a persona-shaped gap you need to close.
Tier defined as: appeared in at least one answer across all 6 buyer personas we primed. Source: 384 AI runs, Apr 2026.
AI doesn't only name agents. For landlords, it names the stack.
When a UK buy-to-let landlord asks ChatGPT how to run their portfolio, six insurance brands, two auction platforms, two software tools and a data provider tie with most estate agents on mention count. This is the adjacent-industry opportunity most agents haven't priced in.
Filtered to non-agent mentions in the BTL Investor persona. Insurance dominates: five of the top twelve are general or landlord-specialist insurers.
The international HNW buyer sees a completely different UK property industry.
When we primed ChatGPT as a Gulf-based HNW buyer looking at Prime Central London, a new tier surfaced — PCL specialists, buying agents, private banks. Most of these firms do not appear in any other persona's answers.
% share of the International Buyer persona's 57 AI runs citing each firm. Beauchamp Estates, John D Wood, Russell Simpson, Rokstone, Property Vision, Prime Purchase and Wetherell appear in no other persona.
The portals answer more than every estate agent brand combined.
total citations of Rightmove, Zoopla and OnTheMarket across 384 AI runs — more than the next 74 visible estate agent brands combined. For anyone trying to be named in AI search, the portals are the gatekeepers.
Citation counts from ChatGPT answers across 384 primed buyer-persona runs, Apr 2026. Firms grouped by canonical name; duplicates merged.
How we built the Index.
We asked ChatGPT 128 questions that real UK property buyers actually search. We ran every question three times, across six primed buyer personas, with live web search forced on every answer. Then we extracted every named UK property firm from every answer, and counted.
PAA + Reddit harvest
Demographic + financial prime
per question / persona
gpt-5.1, forced web search
Landscape reference
In ≥1 AI answer
Full methodology
Question sourcing. 498 unique People-Also-Ask queries harvested across 38 UK property SERPs, scored for buyer-intent and filtered to the top 128 company-surfacing questions. 18 cross-persona questions + 20 persona-specific questions per persona.
Persona priming. Each prompt prefaced with a 2–3 turn prime establishing the persona's demographic, financial situation, buying-journey stage and top concerns before the buyer question lands.
Production run. OpenAI gpt-5.1 via Responses API, forced web_search grounding on every call. 384 LLM calls. Zero errors. $5.95 OpenAI + ~$0.60 extraction.
Firm extraction. Claude Sonnet 4.6 structured JSON extraction. Ranking = count of distinct samples naming each firm per persona.
Landscape reference. 298 UK property firms built from 38 category SERP seeds, DR/traffic-filtered, Claude-classified. This is the invisibility baseline.
Limitations. Single-provider snapshot (ChatGPT only), Q2 2026 timestamp, name-matching edge cases. Full reproduction pipeline in the PDF appendix.
The 6 buyer personas
Downsizer — 58–68, empty-nesters, equity-rich, 12–24 month decision window.
First-Time Buyer — 28–36, dual-income, rate-sensitive, high information appetite.
BTL Investor — 42–58, portfolio-minded, yield-sensitive, tax-aware.
International Buyer — 40–60, HNW overseas, £2M–£15M bracket, PCL-focused.
Relocator — 35–50, family moving for schools, upsizing on space.
Commercial Tenant — 40–58, COO/Head of Ops, 50–500 HC business, lease-driven.
The 60-page Index, in your inbox.
Full per-persona ranking tables for all 25+ firms. The 298-firm visibility matrix, every question we asked, full methodology — and the outreach target list of 221 firms invisible in AI search.
What's in the PDF
- Full 60-page Index — consulting-firm register, data-rich
- Per-persona Top 50 ranking tables · all 6 personas
- 298-firm UK property landscape visibility matrix
- Complete list of the 128 prompts we asked ChatGPT
- Full methodology and reproduction pipeline
- Invisible-firm outreach target list (221 firms)
- Sector deep dives: commercial, property management, retirement
Report Coming Soon
The US index publishes next quarter.
Don't wait. Get a custom report for your domain now.