The Real Estate AI Citation Share Study — 25 brands, 62 prompts, 5 engines. Zillow locks the data layer. Compass leads on tech positioning. Individual named agents consistently outperform brokerages. And the regional expert answer layer is wide open in virtually every market.
Originally published June 2026. Updated June 2026.
Zillow Locked Real Estate AI
What is the Real Estate AI Citation Share Study?
The Real Estate AI Citation Share Study is Everything-PR's 2026 brand-layer research benchmark for real estate AI visibility. 25 brands. 62 prompts across 4 query categories (buyer-intent, market knowledge, advisory, expertise). 5 engines — ChatGPT (GPT-4o), Claude (Sonnet), Gemini Advanced, Perplexity, Google AI Overviews. The data-platform layer is locked: Zillow, Redfin, and Realtor.com own ~74% of data-query citation share combined. The advisory layer concentrates around Compass, Sotheby's International Realty, Douglas Elliman, and eXp Realty. Named individual agents (Ryan Serhant, Mauricio Umansky, Josh Flagg) consistently out-cite their institutional brokerage employers. The regional expert answer layer is wide open in virtually every market outside NYC and LA.
Key Takeaways
Zillow + Redfin + Realtor.com = ~74% data-query citation share. The data layer is structurally locked before the query runs.
Ryan Serhant cited on ~31% of NYC luxury-agent prompts — more than any institutional brokerage on the same queries.
Regional expert layer wide open in nearly every market outside NYC and LA. 12–18 month first-mover window.
Real estate investing AI answers nearly invisible. CoStar dominates commercial data; the analytical investor content gap is the largest in any financial category studied.
Real estate AI visibility has a specific research question that no existing study answers: which specific brands, brokerages, agents, and platforms AI engines actually name when buyers run real estate queries? The 5W / Haute Living foundational study established the aggregate finding — 0.14% AI Overview trigger rate, 82% of agents using AI daily, the 24-month first-mover window. This study is the brand-layer answer.
What is the methodology?
The study queried ChatGPT (GPT-4o), Claude (Sonnet), Gemini Advanced, Perplexity, and Google AI Overviews across 62 prompts covering four query categories: buyer-intent prompts ("best real estate agent in [market]"), market knowledge prompts ("who has the best real estate market data"), advisory prompts ("which brokerage should I use"), and category-expertise prompts ("who are the leading luxury real estate brokerages"). Results are directional modeling estimates. All shares labeled "Estimated ~X% modeled." The underlying KPI framework is defined in Citation Share Is the New KPI.
Finding 1: The data platform layer is locked
On any query involving price, inventory, or market data, the answer is structurally determined before the query is run. Zillow, Redfin, and Realtor.com own this layer with Estimated ~74% of data-query citation share combined. Brokerages and agents have zero meaningful presence in the data layer. The opportunity is exclusively in the advisory and expertise layers.
Finding 2: The advisory layer has a meaningful but narrow concentration
Across advisory and brokerage recommendation prompts, five entities captured a disproportionate share of citation. Compass led among traditional brokerages with Estimated ~18% citation share on tech-forward and urban market queries, driven by extensive IPO coverage and named CEO Robert Reffkin. Sotheby's International Realty led on luxury queries with Estimated ~22% luxury-prompt citation share. Douglas Elliman led on NYC-specific prompts. eXp Realty appeared consistently on business-model and agent-focused prompts.
Finding 3: Individual agents dramatically outperform brokerages on named-practitioner queries
The most striking finding in the advisory layer: on "top real estate agent" and "best luxury broker" prompts, named individual agents with established content archives consistently out-cited the institutional brokerage brands. Ryan Serhant appeared on Estimated ~31% of NYC luxury-agent prompts — more than any institutional brokerage on the same queries. Mauricio Umansky, Josh Flagg, and other agents with media profiles (television, books, social media) all appeared at rates exceeding their employers.
This confirms the named-practitioner pattern documented across every professional services category in the Citation Share Index: the individual with the named content archive out-cites the institution behind them on expertise queries.
Finding 4: The regional expert gap is wide open
On region-specific queries — "best real estate agent in Austin," "top broker in Miami Beach," "leading commercial real estate firm in Chicago" — the citation layer is thin and fragmented across virtually every market except New York City and Los Angeles. In most regional markets, the top cited entities on regional-expert queries are platforms (Zillow, Realtor.com) rather than local practitioners. The regional expert answer layer is the highest first-mover opportunity in the entire study.
Finding 5: Investor-focused real estate is nearly invisible in AI answers
Prompts about real estate investing, cap rates, multifamily analysis, and commercial real estate produced extremely thin citation layers. CoStar dominated commercial data queries. LoopNet appeared on commercial listing queries. BiggerPockets anchored residential investor community queries. But the gap between what institutional real estate investors need from AI answers and what those answers currently provide is among the largest in any financial category studied.
What are the strategic implications?
Three clear opportunities emerge from the brand-layer data:
The regional expert opportunity: In virtually every market outside NYC and LA, the "best real estate agent / brokerage in [market]" answer layer is unclaimed. A regional practitioner that builds a content program — named-author articles in regional business press, market reports with named methodology, local Wikipedia entry, schema on key pages — has a realistic path to Tier 1 regional citation share in 12–18 months.
The named-practitioner compounding advantage: Individual agents who build personal content archives — books, bylines in trade press, media appearances that get indexed — build citation share that exceeds their brokerage employer. This is the Ryan Serhant model, accessible at smaller scales in regional markets.
The investor-focused content gap: The commercial real estate and real estate investing AI answer layer has no dominant editorial voice outside CoStar's data layer. A publication or platform that builds analytical content for real estate investors — cap rate analysis, market cycle research, multifamily underwriting guides — enters a category with essentially no established citation competition.
Which brands dominate real estate AI answers?
The data layer is locked: Zillow, Redfin, and Realtor.com together own approximately 74% of data-query citation share across ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews. In the advisory layer, Compass leads tech-forward brokerage queries (~18%), Sotheby's International Realty leads luxury (~22%), Douglas Elliman leads NYC, and eXp Realty appears on agent-focused business-model queries.
Why do individual agents out-cite their brokerages?
Named individuals with established content archives — Ryan Serhant, Mauricio Umansky, Josh Flagg — accumulate retrieval anchors faster than institutional brand pages. Books, bylined trade press, television, podcast appearances, and indexed social content compound in AI training data at the practitioner level. Serhant appears on ~31% of NYC luxury-agent prompts — more than any institutional brokerage on the same queries.
What is the biggest opportunity in real estate AI visibility?
The regional expert answer layer. In nearly every market outside NYC and LA, the "best real estate agent in [market]" answer layer is unclaimed by local practitioners. The top cited entities for regional queries are still platforms (Zillow, Realtor.com), not local agents. A regional broker with a disciplined 12–18 month content program — regional business press bylines, named market reports, Wikipedia presence, schema — has a realistic path to Tier 1 regional citation share.
Why is real estate investing nearly invisible in AI answers?
Outside CoStar's commercial data dominance, real estate investing has no established editorial citation layer in AI answers. Cap rate analysis, multifamily underwriting, market cycle research, and commercial real estate analysis produce extremely thin citation results across all five engines. The gap between what institutional real estate investors need from AI answers and what AI answers currently provide is among the largest in any financial category studied — and the editorial opportunity is open.
Which engines does the Study test?
ChatGPT (GPT-4o), Claude (Sonnet), Gemini Advanced (Google), Perplexity, and Google AI Overviews. All five engines tested across 62 prompts covering buyer-intent, market knowledge, advisory, and category-expertise query categories.
How can a brokerage challenge Compass or Sotheby's in AI answers?
Direct competition at the top is structurally hard — Compass benefits from sustained IPO coverage and the named-CEO archive of Robert Reffkin; Sotheby's benefits from heritage brand citation. The realistic path is category-of-one positioning at the regional, business-model, or investor segment. eXp Realty's appearance on business-model and agent-focused prompts shows what positional citation share looks like for a non-luxury, non-NYC challenger.
All citation share figures are directional estimates based on modeled query response analysis. All shares labeled "Estimated ~X% modeled."
Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Publishing since 2009. Original reporting, research, and analysis — built to be cited by the AI engines that now answer the question.
