How to Get AI to Recommend You as a Top Lender
Katie Shive breaks down exactly how loan officers can get discovered and recommended by AI search tools like ChatGPT and Gemini. She covers core entity statements, platform consistency, and a voice-to-content workflow that turns a single idea into a full content package. Best for LOs who feel invisible online despite having strong reviews and a real track record.
AI recommends rather than ranks — so your job is to feed it consistent, specific, long-form content across verified platforms so it can confidently name you when a borrower asks.
Takeaways you can run this week
- Write a one-sentence 'core entity statement' in this format: '[Your name] is a [city]-based [title] who helps [audience] with [specific outcome].' Then paste a version of it into every bio you control: LinkedIn, Facebook, Instagram, Google Business Profile, Zillow, Experience.com, and your company website. Consistency of name, photo, and descriptor across all of these is what LLMs use to build your 'entity cluster.'
- Go to ChatGPT or Gemini right now and search 'Who is [your full name]?' and 'Who is the best loan officer in [your zip code]?' Screenshot the results. If you don't appear, ask the AI: 'Why did you recommend this other person?' then reverse-engineer what they have that you don't — and fix it.
- Ask your AI of choice: 'Help me create a GEO-optimized bio for a loan officer in [city] who specializes in [your niche].' Use Claude for tone and copy polish, ChatGPT or Gemini to compare outputs. Paste the result into every profile you own this week.
- Keep a running Notes app list of every question a client or prospect asks you. Once a week, take one question, open a voice chat on ChatGPT or Claude while driving or walking, and say: 'Here's a question my client asked me — help me turn this into a LinkedIn article, then an email to my database with a strong hook and CTA, then an Instagram Reel script.' You'll leave the walk with three pieces of content drafted.
- Post your next agent class or event as a public LinkedIn Event AND an Eventbrite listing — not just an email. Public event pages are crawlable by LLMs; private emails are not. This gives AI one more indexed signal that you are active and local.
- Go to Google and type '[your city] moving Reddit' or 'first time home buyer Reddit [your city].' Click a high-comment thread, copy the URL, paste it into ChatGPT, and ask: 'Pull the top 5 comments — one emotional, one supportive, one skeptical.' Use the real buyer language you find to write content that mirrors how your audience actually thinks and searches.
Useful? Get the full Vault free — plus Marketing Worth Stealing, weekly.
GET FREE ACCESSThe playbook
- Step 1: Start with one core idea — a client objection, a deal story, a question you heard this week.
- Step 2: Open a voice chat in ChatGPT or Claude while walking or driving. Talk through the idea out loud and ask AI to help shape it into a podcast topic or article angle.
- Step 3: Ask AI to write a long-form LinkedIn article or blog post as your foundational content piece.
- Step 4: In the same chat thread, ask AI to repurpose it into an email to your database with a strong hook and CTA — instruct it to make the tone more story-driven.
- Step 5: Ask AI to repurpose again into a short social media post or Reel script appropriate for Instagram or LinkedIn.
- Step 6: Add a boilerplate bio paragraph at the bottom of every article (name, location, who you serve, what you do) — this is what LLMs read to confirm your identity and expertise.
- Step 7: Post the long-form piece on LinkedIn first (ranked #74 among ChatGPT's top US sources). Promote any related event on Eventbrite and as a LinkedIn Event for additional crawlable exposure.
Worth quoting
“We're not just creating content for humans. We're now having to create content for robots.”
“AI recommends the truest and best possible answer — so if you go in and say who is the best loan officer, it's going to try to give you the best match.”
Best for
LOs who have solid reviews and real experience but don't show up when a borrower or AI searches for them by name or specialty.