GEO case study

How we ranked Scott Evansin ChatGPT’s top answers.

Scott is a purchase-focused loan officer in San Diego who has built his business on agent partnerships since 2006. We made his specialty impossible for AI assistants to miss.

Scott Evans

Scott Evans

Executive Vice President

CrossCountry Mortgage · San Diego, CA

LinkedIn profile
Market
San Diego, Chula Vista, Oceanside, La Mesa
Specialty
Purchase, conventional, FHA, VA and jumbo loans
2025 volume
$398.4M across 641 loans (avg. ~$621K)
NMLS
256163

The starting point

Scott closes more loans than almost anyone in Southern California, most of them purchases through agents. But his online presence was split between a company bio, a personal LinkedIn and a separate LinkedIn business page. Buyers asking ChatGPT about VA loans in San Diego or a purchase lender agents trust got generic answers with no local loan officer in them.

The GEO strategy we ran

1. A website built around his specialties

His Homes AVM website leads with what he does best: purchase loans, VA, FHA and jumbo financing and fast pre-approvals for agents’ buyers. Person, FinancialService and FAQPage schema make that clear to crawlers, with sameAs links to both LinkedIn pages and his CrossCountry Mortgage profile.

2. Answer-first guides for San Diego buyers

  • VA loans in San Diego: eligibility, funding fee and county loan limits
  • Jumbo loans in San Diego County: what buyers need in 2026
  • First-time buyer programs in San Diego County
  • FHA vs conventional for San Diego condos
  • How fast can I get pre-approved in San Diego

3. Listings that match everywhere

One consistent profile across Google Business Profile, Bing Places, Apple Business Connect, Zillow, Yelp and NMLS Consumer Access, with the same name, NMLS number, phone and service area.

4. Backlinks from the people buyers trust

Scott already teaches agents about prospecting and team building. We turned that into links and mentions from agent partner websites, training sessions and community event pages, exactly the sources AI assistants cite for local recommendations.

5. Reviews that say the loan and the neighborhood

Homes AVM review requests ask clients to mention their loan type and neighborhood, building a body of reviews that matches the questions buyers ask.

6. Home reports that bring clients back

Every past client gets a monthly Homes AVM home report, which drives repeat visits, referrals and new reviews, keeping his profile active and current.

The result

When someone asks ChatGPT for a top loan officer for purchase or VA loans in San Diego, Scott is now in the answer, described as a high-volume purchase specialist trusted by local agents. We track those prompts monthly across ChatGPT, Perplexity, Gemini and Google AI Overviews.

Be the name ChatGPT suggests in your market.

Your Homes AVM website, listings and monthly home reports work together so AI assistants can find you, trust you and recommend you.