GEO FOR HEALTHCARE: HOW TO SHOW UP WHEN PATIENTS ASK AI FOR CARE

A patient is diagnosed with trigeminal neuralgia. Their doctor mentions surgery might help. That night, they open ChatGPT and type: "best neurosurgeon for trigeminal neuralgia near me." The AI responds with a list of Healthgrades profiles, a Zocdoc link, and a WebMD overview of the condition. The neurosurgeon with 2,000 successful cases and a 98% satisfaction rate? Nowhere in the answer.

This is happening right now, thousands of times a day, across every specialty. And it is about to get worse for providers who do not adapt.

THE SHIFT: AI IS THE NEW FIRST FILTER

73% of patients research online before choosing a provider. That number has held steady for years. What has changed is where they start. Increasingly, patients are skipping Google entirely and going straight to ChatGPT, Google AI Overviews, or Perplexity with questions like "where can I get radiosurgery in [city]" or "best orthopedic surgeon for ACL repair near me."

AI does not return ten blue links. It returns one answer. A paragraph naming specific providers, clinics, or (more often) third-party directories. If your practice is not in that paragraph, the patient never sees you. They never visit your website. They never call your office. You were filtered out before the search even felt like a search.

This is Generative Engine Optimization (GEO) applied to healthcare. And right now, almost no healthcare organization is doing it.

WHAT PATIENTS SEE TODAY

Ask ChatGPT for a specialist in any city. Try it right now. You will notice a pattern: the AI recommends Healthgrades profiles, Zocdoc listings, and WebMD condition pages. It rarely links to the physician's own website or the hospital system's provider page.

This means third parties own your digital identity unless you actively claim it. The doctor with 30 years of experience, 50 peer-reviewed publications, and outcomes data that puts them in the top 5% nationally? AI does not know any of that. It knows what Healthgrades says, because Healthgrades has structured data, schema markup, and content that AI can parse. The doctor's own website, built on a legacy CMS with no structured data, is invisible.

The irony is painful. The providers with the best clinical outcomes often have the worst digital visibility. Their expertise lives in journals, conference presentations, and patient outcomes databases that AI tools cannot access or cite.

WHY HEALTHCARE SITES SCORE POORLY

We have audited healthcare organizations from solo practices to major hospital systems. The patterns are consistent:

• No MedicalOrganization or Physician schema. These are the exact schema types that AI tools use to understand healthcare entities. Without them, AI is guessing that your site is even a medical practice, let alone what specialties you offer or what conditions you treat.
• No condition-specific landing pages. Patients search by condition, not by doctor name. They type "Gamma Knife treatment for brain metastases" not "Dr. Smith neurosurgeon." If you do not have a dedicated page for each condition you treat, written for patients, you do not exist for that search.
• No llms.txt telling AI how to describe the practice. This plain text file takes 30 minutes to create and directly instructs AI models on your specialties, your providers, and your clinical focus. Almost zero healthcare organizations have one.
• Provider directories that are JavaScript-rendered. Many hospital systems use React or Angular to render their "Find a Doctor" pages. These pages look fine in a browser, but AI crawlers see an empty page. Every provider in that directory is invisible to AI search.
• Legacy CMS platforms that cannot support modern structured data. Healthcare organizations often run on enterprise CMS platforms that were deployed 8 to 12 years ago. Adding schema markup, creating llms.txt files, or restructuring content for AI readability requires workarounds that the platform was never designed to support.

THE 5 THINGS EVERY HEALTHCARE WEBSITE NEEDS

1. Physician schema for every provider

Every doctor on your website should have Physician schema markup that includes their name, specialties, conditions treated, medical credentials, affiliated hospitals, and accepted insurance networks. This is how AI tools build their internal knowledge graph of who treats what. Without Physician schema, your doctors are just names on a page.

2. MedicalOrganization schema for the practice or hospital

Your organization needs MedicalOrganization schema that declares what you are, where you are, what specialties you offer, and how patients can reach you. This is the structural foundation that tells AI "this is a medical facility that treats these conditions in this location." Combined with Physician schema, it creates a complete picture AI can cite with confidence.

3. Condition-specific pages, one per condition treated

This is the single highest-impact change most healthcare organizations can make. Create a dedicated page for every condition your practice treats. Write it for patients, not for physicians. Include symptoms, treatment options you offer, outcomes data if available, and a clear path to schedule. When a patient asks AI about a condition, the AI needs a specific, authoritative page to cite. Generic "Our Services" pages do not qualify.

4. An llms.txt file

A plain text file at yourdomain.com/llms.txt that gives AI a structured briefing on your organization. For healthcare, this should include your clinical specialties, key providers and their credentials, conditions treated, technologies or techniques used, locations, and any notable outcomes or certifications. Here is exactly how to create one.

5. Google Business Profile optimization

Google Business Profile remains the anchor of local healthcare search. When AI tools answer location-based healthcare queries, they pull heavily from Google's local data. Your GBP needs to be complete, accurate, and active: correct categories, service areas, photos, reviews being responded to, and Q&A populated with real patient questions. For multi-location health systems, every location needs its own optimized profile.

THE SPECIALTY DIMENSION: WHAT MAKES HEALTHCARE DIFFERENT

In B2B, you audit the whole site as one entity. A manufacturer's AI visibility score applies across their business. Healthcare is fundamentally different. You need to score visibility per clinical indication.

A neurosurgeon might show up when a patient asks about brain tumors but be completely invisible when they ask about trigeminal neuralgia, even though they treat both conditions daily. A cancer center might appear for breast cancer queries but not for the rare sarcoma program that is actually their differentiator.

This is why a healthcare AI search visibility audit requires an indication matrix: a condition-by-condition breakdown of where the provider or system appears (and where it does not) across AI search tools. The overall score tells you there is a problem. The indication matrix tells you exactly where.

Without this, you are optimizing blind. You might invest in content for conditions where you already rank while the conditions that drive the most patient volume remain completely invisible.

REAL EXAMPLES

THE WINDOW IS OPEN

Healthcare is behind every other industry in AI search optimization. That is both the problem and the opportunity. The first practice or hospital system in a metro area to implement Physician schema, condition-specific pages, and an llms.txt file will dominate AI recommendations for their specialty. AI tools develop preferences for sources they can parse and cite reliably. Once you are established as the go-to source, competitors have to outwork you to displace you.

Patients are already asking AI for care recommendations. The only question is whether they find you or your competitors. Just like in manufacturing, the organizations that move first will be the ones AI remembers.