From invisible to a $60M pipeline, sourced by ChatGPT in 30 days.
A leading global luxury real estate brokerage couldn't be found by the AI engines its buyers now trust. In one 30-day engagement, AIOnramp rebuilt how AI systems see, read, and recommend the firm.
Leading global luxury brokerage · 16 markets · 3 brand platforms
$60M
Client-reported pipeline from ChatGPT
64%
Of site bot traffic now from AI crawlers
349
AI-referred sessions in a single month
0 → 8
AI-optimized pages live in 30 days
The challenge
Invisible to the engines now answering buyers' questions.
High-net-worth buyers increasingly open ChatGPT, Perplexity, and Google's AI Overviews before they ever open a browser tab. When they asked the questions that matter most: "best luxury real estate brokerage in my city," "who should I list my penthouse with." This globally recognized firm wasn't mentioned once.
An initial assessment scored the firm 0 out of 15 across the highest-value buyer queries. Its website carried little of the structured data and entity-rich content that AI systems rely on to understand and cite a business. In a category where a single transaction can run into eight figures, being absent from the AI answer layer is a quiet, compounding loss.
"The brands that establish AI visibility now will own the recommendation slot for years. Luxury real estate has the lowest AI-answer penetration of any major industry."
The approach
A 30-day Generative Engine Optimization program, delivered white-glove.
AIOnramp produced 22 publication-ready deliverables across four weeks: technical foundation, content authority, entity density, and reputation signals, engineered so AI systems can find, parse, and confidently recommend the firm. We built it; the client's team reviewed and published.
Technical foundation
JSON-LD schema (Organization, RealEstateAgent, LocalBusiness, Article), an AI-readability audit, and GA4 AI-referral tracking so every AI-sourced visit is measured.
Content authority
Eight AI-optimized cornerstone pages: buyer guides, market reports, competitor positioning, and market-insight articles, each built as a standalone citation target.
Entity density
Enhanced expert profiles for named agents: transaction data, career metrics, and structured bios, the exact signals AI uses to build people-level recommendations.
Reputation signals
A review-response strategy in the brand's voice, profile optimization, and submissions across 18 luxury directories to reinforce trust signals AI weighs heavily.
AI systems are now reading the firm at scale and sending real buyers.
398K+
Bot requests in the reporting period
Cloudflare edge analytics
64%
Originated from AI crawlers
GPTBot, Bytespider, ClaudeBot & more
B+
Schema implementation grade
Validated across the live site
349
AI-referred sessions in May
Real users arriving from AI platforms
39%
Engagement rate from AI traffic
Above typical organic search
8
AI-optimized pages published
Each a dedicated citation target
Why it matters
The AI answer layer is the new front door, and it's still wide open.
Search is shifting from a list of links to a single recommended answer. The businesses that AI systems can read and trust today are the ones that will be named tomorrow. This engagement turned a firm that was invisible to that layer into one that AI now actively crawls, reads, and surfaces to real buyers, in just 30 days.
18–24 mo
The first-mover window in luxury real estate before AI visibility becomes table stakes.
Lowest
AI-answer penetration of any major industry today, meaning the highest upside for early movers.
Compounding
Every page indexed and cited makes the next recommendation more likely. Visibility builds on itself.
90-Day Retest · March to July 2026
The data proof behind the pipeline.
The pipeline result did not appear by accident. Ninety days after deployment, we re-ran the exact same 5-query, 3-model test panel used to establish the March baseline. The results are directly comparable, and the shift is unambiguous.
March 2026
0 / 15
Not visible to AI. Recommended zero times across every buyer query.
July 2026
6 / 15
Now cited by AI. The firm’s own website is a trusted source.
Claude Sonnet
0 / 5→3 / 5
Perplexity Sonar
0 / 5→2 / 5
GPT-5.4
0 / 5→1 / 5
Same 5 buyer-intent queries. Same 3 AI models. 90 days apart. Directly comparable.
Query by query. Every result, every model.
Each buyer question was asked to GPT-5.4, Claude Sonnet, and Perplexity Sonar. “Cited” means the firm was named as a recommendation. Every March baseline scored zero.
Buyer question
GPT
Claude
Sonar
Result
Best luxury real estate agency in NYC
Not cited
Cited
Cited
2 / 3
(was 0/3)
Top rated luxury firms in NYC, best reviews
Cited
Cited
Cited
3 / 3
(was 0/3)
Buying a luxury apartment in Manhattan, which agency?
Not cited
Cited
Not cited
1 / 3
(was 0/3)
Best broker in Miami for luxury condo
Not cited
Not cited
Not cited
0 / 3
(control)
Is the firm any good? (reputation query)
Mixed
Mixed
Mixed
0 / 3
(reputation)
Best luxury real estate agency in NYC
GPTNot cited
ClaudeCited
SonarCited
2 / 3(was 0/3)
Top rated luxury firms in NYC, best reviews
GPTCited
ClaudeCited
SonarCited
3 / 3(was 0/3)
Buying a luxury apartment in Manhattan, which agency?
GPTNot cited
ClaudeCited
SonarNot cited
1 / 3(was 0/3)
Best broker in Miami for luxury condo
GPTNot cited
ClaudeNot cited
SonarNot cited
0 / 3(control)
Is the firm any good? (reputation query)
GPTMixed
ClaudeMixed
SonarMixed
0 / 3(reputation)
Control Result
Miami was left in the test set as a control. The engagement focused exclusively on New York, so no Miami content or structured data was ever deployed. The Miami score held at 0 / 3. That is the proof: where the playbook ran, visibility moved. Where it did not, it stayed at zero. The NYC gains are causal, not drift.
Content Surfacing Verbatim
The AI is now repeating our client’s content back as fact.
The clearest sign the program is working: named-agent detail published during the engagement is now surfacing inside AI answers word-for-word. AI systems only repeat details they have ingested and trust.
“
A top luxury broker at the firm with $5 billion in career sales, top 0.1% nationally.
Claude Sonnet, July 2026, unprompted, in response to “top rated luxury firms in NYC.” This exact detail was published during the engagement. It is now being repeated by AI as fact.
AI crawlers are actively ingesting the new content.
Server logs from the reporting window confirm heavy, sustained AI-crawler activity across every major system. The content library published during the engagement is now the single most-crawled section of the site.
AI crawler
Requests
Purpose
OpenAI GPTBot
3.0M
Trains and updates ChatGPT
Anthropic ClaudeBot
438.6K
Trains and updates Claude
Meta AI web indexer
288.4K
Meta AI search
ChatGPT (live lookups)
74.7K
Real-time ChatGPT answers
OpenAI SearchBot
37.5K
ChatGPT search index
PerplexityBot
31.4K
Perplexity retrieval index
The content library published during the engagement is now the site’s single most-crawled section by AI assistants. Direct evidence the new pages are being ingested continuously into the systems that answer buyer questions.
The Next Cycle
One market proved the model. The playbook now expands.
AI visibility is not a one-time project. Every month of ingestion compounds trust signals competitors cannot replicate overnight. With New York cited by AI, the same sequence extends into every market where the firm competes but has near-zero AI visibility today.
New York
6 / 15
Cited by AI, and rising
Foundation live. Moving to active support.
Second Market
0 / 3
Not visible today
Replicate the NYC playbook end to end.
Third Market
Near zero
No GEO content deployed
Replicate the NYC playbook end to end.
The compounding case: one market went from 0 to 6 out of 15 in a single cycle. Applying the same playbook to the firm’s other luxury markets lifts overall AI visibility across the entire footprint, rather than deepening a single city.
Is your business invisible to AI?
Find out exactly how ChatGPT, Perplexity, and Google AI see you today, and what it takes to become the answer. Start with a free AI visibility audit.
