AEO Case Study
From no website to top ChatGPT results for a high-intent real estate query.
This client was a real estate broker targeting one exact commercial prompt in ChatGPT: [nationality] real estate agent [state]. Before the engagement, there was no website and the client did not appear at all. During the February 2026 engagement, the client began appearing in the top ChatGPT results for that query within 2 to 3 weeks of launch.
Snapshot
Engagement
February 2026
The work behind this case study was completed during a February 2026 client engagement.
Client Type
Real estate broker
A local broker wanted to appear for a nationality-plus-state real estate query in ChatGPT.
Starting Point
No website, no visibility
There was no site in place and the client did not appear at all for the target prompt before the work started.
Timeframe
2 to 3 weeks
The published screenshot reflects the client appearing in the top ChatGPT results within roughly three weeks.
Published Result
Top ChatGPT results
We are publishing the result anonymously, but the screenshot on this page is from the real client outcome.
The prompt we optimized for
We focused the work around one narrow, local, high-intent query instead of trying to rank for every possible real estate phrase at once.
[nationality] real estate agent [state]That kept the strategy clear: make the broker easy to identify, easy to verify, and easy to cite for exactly that search pattern.
What Changed
Built a site around one exact commercial prompt
The project started from zero, so the site could be organized directly around the query pattern the client cared about most.
Added targeted metadata and on-page entity signals
Title tags, descriptions, geo signals, language signals, and service-language combinations were all aligned to the target market and geography.
Published structured data across the key entity types
The implementation included RealEstateAgent, LocalBusiness, FAQPage, WebSite, Organization, and BreadcrumbList schema to make the business easier to parse and verify.
Created direct-answer content
The site included FAQ content, service sections, area-served coverage, and language-specific context that mapped to how people phrase local AI prompts.
Shipped AI-readable and crawler-facing files
We added llms.txt, llms-full.txt, robots.txt, and sitemap.xml so the site was easier for answer engines and crawlers to fetch and interpret.
Strengthened corroboration
The entity layer linked out to established third-party profiles and credentials so the on-site claims were supported by outside references.
Published Result
This is the same screenshot used on the homepage. We are keeping the client anonymous, but the result is real and tied to the prompt above.
Why We Think It Moved
Clearer entity matching
The business, language, geography, and profession were described consistently across metadata, visible copy, and schema.
Better retrieval surfaces
The site answered the query directly instead of forcing a model to infer the match from thin or generic content.
Richer machine-readable signals
Structured data, FAQ markup, and AI-readable files gave answer engines more ways to interpret the business correctly.
Outside confirmation
Linked profiles, credentials, and organizational context helped support the trust layer beyond the site itself.
FAQ
What was the starting point for this client?
The client was a real estate broker targeting a very specific nationality-plus-state query in ChatGPT. At the start of the engagement there was no website and the client did not appear in results for the target prompt.
What prompt was the work built around?
The primary prompt was a nationality-plus-state real estate agent query. We centered the site structure and machine-readable signals around that high-intent query pattern while keeping the client anonymous in public-facing materials.
What changes did AI NYC make?
We built the site from scratch, added targeted metadata, shipped multiple schema types including RealEstateAgent and FAQPage, created FAQ and area-served content, published llms.txt and llms-full.txt, exposed crawler-friendly sitemap and robots files, and strengthened entity corroboration through profile links and trust signals.
How long did it take to see movement?
During the February 2026 engagement, the client began appearing in the top ChatGPT results for the target prompt within roughly 2 to 3 weeks. Timelines vary by market, competition, and prompt volatility.
Does this mean AI NYC guarantees the same result for every client?
No. AEO is still an emerging field and AI model behavior can change without notice. This case study is published as an anonymized example of a real outcome, not as a guarantee.
Start with the free audit, then see how the same principles apply to your market.
This case study is one anonymized example, not a universal template. If you want to know where your own site stands, start with the audit and then review the NYC service page for the broader commercial model.