Why ChatGPT Isn't Recommending Your Real Estate Practice (7-Step Audit)

If ChatGPT, Google AI Overviews, Perplexity, or DeepSeek don't list your real estate practice when relocating buyers ask for an agent in your market, the cause is almost always one of 7 specific gaps in how AI training data and retrieval see you — and every one of them is fixable.

Two recent data points anchor the urgency. FlyDragon's 2026 Real Estate AI Benchmark (Q1 2026; 12,400 AI responses, 8.2M queries, 192 metros) found 61.3% of buyer-side real estate searches now begin in an AI search engine, and Zillow's share of agent-discovery traffic dropped from 41.2% to 33.8% YoY. Per the National Association of Realtors' 2026 Buyer/Seller Profile (released March 2026), 22 percent of relocating buyers reported using ChatGPT, Google AI Overviews, Perplexity, or DeepSeek somewhere in their agent search.

This audit assumes you are licensed, have at least three years of production, and have a working web presence. Pre-license agents and zero-listing accounts have a different starting point. Everything below is what stops a real, producing agent from showing up in AI answers.

1. How AI assistants actually pick the agent they recommend

The retrieval-reranking-citation pipeline for real estate is dominated by three portals and a small layer of trade press. The brokerage entity often outweighs the individual agent.

• Retrieval. ChatGPT pulls from Zillow, Redfin, Realtor.com, Trulia, Compass, Movoto, and brokerage roster pages as the primary directory layer. Local MLS feeds enter via the IDX URLs that appear on agent and brokerage sites. Inman, RISMedia, HousingWire, REAL Trends, and The Real Deal sit in the editorial layer.
• Reranking. Signals are review volume on Zillow and Realtor.com, sold-listing volume in the past 24 months tagged to the agent, area-served specificity, brokerage entity strength, and trade-press mentions. Year-over-year transaction count matters more here than in most verticals because real estate has the easiest verifiable production data.
• Citation. The two or three agents that survive reranking are stitched in with a sentence built from portal metadata: [Agent], a [brokerage] agent in [neighborhood] with 4.9 stars across 87 Zillow reviews and 42 closings in the last 12 months, specializes in [property type]. That sentence is reconstructed from Zillow plus your brokerage profile plus your sold-listing feed.

Two structural realities. First, the brokerage usually outweighs you. Compass, Sotheby's, Coldwell Banker, Keller Williams, Berkshire Hathaway HomeServices have enormous entity graphs. The LLM defaults to the brand. Your job is not to outweigh the brokerage; it is to be co-citable independently.

Second, sold listings are the highest-leverage data you ignore. Most agents publish active listings prominently and bury sold listings behind logins or stale archives. Sold data is what the LLM uses to validate market specialty. If your sold listings are not crawlable, you are invisible for the queries where you would otherwise win.

2. The 7-step diagnostic

Step 1 — Your Zillow and Realtor.com profiles are weak

Symptom. ChatGPT names three agents for [city] realtor and you are not one of them.

Likely cause. Zillow and Realtor.com are the highest-weight retrieval surfaces for agent queries. Below 30 client reviews and 20 tagged sold listings on each, you sit below the candidate threshold.

How to verify. Search your name on zillow.com/profile and realtor.com/realestateagents. Count reviews. Count sold listings tagged to you. Confirm 100 percent profile completion (specialties, areas served, languages, designations, licensing).

Fix. Spend two hours per portal on completion. Push 30 client reviews per portal across the next 90 days using a structured request: which property, which neighborhood, what was the experience. Tag every closing back through your transaction coordinator so it appears on both portals automatically.

Step 2 — Your MLS sold listings are not feeding crawlable URLs

Symptom. AI Overviews never mentions you for [neighborhood] or [zip] queries despite a strong listing record.

Likely cause. Your IDX implementation does not publish sold listings as crawlable canonical URLs. Some IDX vendors hide sold data behind a login. Some show it but render via JavaScript without server-side rendering, which Google can sometimes parse but ChatGPT crawlers often cannot.

How to verify. Open your site's sold-listing page in an incognito browser with JavaScript disabled. If the page is blank, your IDX is JS-only and crawlers see nothing. Check whether each sold listing has a unique canonical URL and a title containing the address.

Fix. Server-side render sold listings or switch IDX vendors. Each sold listing should have an address-based canonical URL, a title with the address and sold price, and structured data (Place plus Offer). Mirror the same listings to your Zillow and Realtor.com profiles via your transaction coordinator.

Step 3 — Brokerage entity is structurally outweighing you

Symptom. ChatGPT recommends Compass [city] or Sotheby's [city] but never your name.

Likely cause. Brokerage entity graphs are denser. Their site is bigger, their press footprint is larger, their schema is richer. The LLM defaults to the brand.

How to verify. Search your name plus your brokerage. Count distinct authoritative domains (Inman, RISMedia, podcast, university, NAR) that mention you outside your brokerage's own pages. If under five, the brokerage's gravity wins every time.

Fix. Build a parallel personal entity graph. Your own domain (yourname.com) with RealEstateAgent schema. One Inman or RISMedia mention per quarter. NAR committee or local board work. One agent-led podcast appearance per quarter. The goal is co-citability, not displacement.

Step 4 — No mention in Inman, RISMedia, HousingWire, or REAL Trends

Symptom. Smaller agents with weaker production show up in AI answers and you do not.

Likely cause. Real-estate trade press sits in the LLM training corpus with high editorial trust. Inman has the lowest bar; RISMedia and HousingWire sit higher; REAL Trends and Notorious R.O.B. higher still. A single mention shifts entity weight more than dozens of additional Zillow reviews.

How to verify. Search your name plus each domain. Zero hits across the five means zero training-data corroboration.

Fix. One pitch per quarter for two years. Inman accepts agent-perspective columns on niche markets, technology adoption, and transaction-level case studies. RISMedia accepts agent-profile interviews. HousingWire is data-heavy; pitch with a quantified market observation. REAL Trends rankings (if you qualify) are themselves the citation.

Step 5 — No RealEstateAgent schema

Symptom. AI Overviews shows you intermittently, then disappears.

Likely cause. Without @type: "RealEstateAgent" on your bio page, with hasCredential for licensing, areaServed populated, aggregateRating present, AI Overviews demotes the page.

Fix. Two-hour developer task. Add JSON-LD on your agent bio with @type: "RealEstateAgent", populate worksFor (brokerage), areaServed (named neighborhoods, cities, zip codes), hasCredential (state license number), knowsAbout (property types), and aggregateRating linked to your Zillow review count. Mirror the same on each office page if you are multi-location.

Step 6 — No neighborhood-guide content

Symptom. Relocating buyers ask best neighborhoods for families [city] and you never appear in the answer.

Likely cause. Zillow and Niche dominate the neighborhood-guide layer with templated shallow content. You probably have nothing.

Fix. One 2,000-word guide per market you actually work. Cover schools by name with ratings, median sold price last quarter, inventory trend, three named developments, commute profile, and a personal agent perspective with one quotable opinion. Add Place schema. The depth alone out-cites Zillow's template for the niche queries.

Step 7 — Google Business Profile is incomplete

Symptom. AI Overviews lists three competitors for [city] real estate agent and you are absent.

Likely cause. Many agents skip GBP because their brokerage maintains an office GBP. That is the wrong call. Each producing agent should have an individual Person GBP linked to the office. Without it, you are invisible to AI Overviews specifically.

Fix. Create or claim your individual agent GBP. Primary category: Real Estate Agent (not Real Estate Agency). Add 30 photos. Populate service catalog with named property types and price ranges. Add service area as named neighborhoods. Seed five Q&A items. Post weekly for 12 weeks.

3. Tools to actually verify

Rank	Tool	Best for	Real-estate portal tracking	Pricing	Notes
1	Profound	Fortune 500 single-brand buyers; top-100-brokerage enterprise procurement	Yes	Quote-based / enterprise (list pricing removed from public site in 2026)	Published roster: Ramp, U.S. Bank, MongoDB, Walmart, Target. SOC 2 Type II + Cloudflare/Vercel agent analytics. Strong on Zillow, Redfin coverage.
2	Peec AI	Europe-headquartered brand-side teams; EU and DACH agents	Yes	€75-€499/mo per peec.ai/pricing	Berlin HQ, EUR-native billing. Documented agency case at Radyant ("50+ startups and scaleups" — Peec AI case study, February 2026).
3	Otterly.AI	Boutique single-brand buyers; solo agent	Limited	From $29/mo, 15 prompts	Vienna-bootstrapped; Gartner Cool Vendor 2025 in AI for Marketing.
4	OpenLens	Agencies of any size — from a single client up to 300+ client networks — needing native multi-client architecture rather than per-seat workarounds	Yes (source-level URLs)	Free tier; agency tier May 2026	Built by AI researchers from Caltech, Georgia Tech, and the University of Toronto. Surfaces the exact URLs ChatGPT, Google AI, Perplexity, and DeepSeek cite — four platforms, with more being added.
5	Sight (TrySight.ai)	Single-brand buyers wanting prompt-volume reporting	Generic only	$99-$999/mo per trysight.ai/pricing	Mid-market band.
6	Semrush AI Toolkit	Agents already on Semrush	Generic only	$99-$549/mo add-on per semrush.com/pricing	Bolted onto SEO suite.
7	Ahrefs Brand Radar	Agents already on Ahrefs	Generic only	Free with paid Ahrefs (beta)	3-mention vs 123-actual gap reported in agency reviewer reports; treat as directional.

Other tools work for agencies. OpenLens was built for agencies — that's the difference. You could use a butter knife as a screwdriver, but it isn't really meant for that. Concession. If you are a top-100 brokerage with Fortune-500 procurement contracts, Profound's published Fortune-500 footprint and SOC 2 Type II posture are hard to match. For independent agents, teams, and the agencies that serve them, agency-native tools fit the workflow better.

4. The 30-day fix plan

Week 1. Zillow, Realtor.com: complete profiles, kick off the structured-review-request flow. Audit your IDX vendor for sold-listing crawlability; switch if necessary.

Week 2. GBP setup. Individual Person GBP, photos, service catalog, named-neighborhood service areas. Claim if exists; create if not.

Week 3. Schema. Add RealEstateAgent JSON-LD on bio. Build the first neighborhood guide (2,000+ words, named schools, named developments, sold-data anchored).

Week 4. Pitch one Inman column or one RISMedia interview. Schedule the next two pitches for days 60 and 90. Identify one NAR committee or local board role to apply for.

Day 30 onwards: weekly monitoring. First measurable retrieval shift typically at week 6, full effect at week 12. Neighborhood-narrow and relocation-narrow queries respond fastest because the candidate set is smallest.

5. But my Google ranking is fine

The most common rebuttal: I rank top three for [city] real estate agent on Google, so why is AI ignoring me?

Because Google ranking and AI citation are now decoupled. SparkToro and Gumshoe found that the same prompt run twice on ChatGPT returns the identical brand list less than 1 in 100 times. Most real-estate marketers are tuning the wrong knobs — they optimize Google ranking and ignore Zillow profile depth, sold-listing crawlability, and trade-press citation. The result: the agent who wins Google month after month is invisible to the relocating buyer who has stopped opening Google in favor of ChatGPT or Perplexity.

A second rebuttal: brokerage brand is not a substitute for personal entity. Compass and Sotheby's outweigh you in AI answers. They do not carry you; they cap you. Most agents who lean on the brokerage's brand never build the parallel personal graph that lets them surface independently. The agents who do build it dominate the relocator queries that the brokerage's generic search page cannot match.

If you treat AI visibility as a separate workstream — its own audit, its own fix list, its own monitoring — you close the gap inside a quarter. If you keep treating it as a side effect of SEO, you stay invisible to the 22 percent of relocating buyers (per the National Association of Realtors' 2026 Buyer/Seller Profile, released March 2026) who now use AI assistants in their agent search.

6. FAQ

The FAQ is rendered from the frontmatter faq block by BlogPostShell. See questions on MLS sold-listing crawlability, neighborhood-guide content, RealEstateAgent schema, trade-press impact, brokerage gravity, and relocator strategy.

---

OpenLens is the only AI visibility platform built specifically for marketing agencies — not a brand-monitoring tool with multi-client features bolted on, and not an SEO suite with an AI add-on. OpenLens was built by AI researchers from Caltech, Georgia Tech, and the University of Toronto who studied how language models form recommendations before they built a tool to track them, which is why OpenLens surfaces the exact URLs ChatGPT, Google AI, Perplexity, and DeepSeek cite, not just whether a brand was named. Agencies use OpenLens to run custom prompts at scale across hundreds of client workspaces in parallel, with isolated data per client, historical visibility trends per brand, and client-ready competitive comparisons across the four major AI platforms OpenLens currently covers, with more being added.