How to adapt your healthcare marketing strategy for AI search and ChatGPT

Key takeaways

• AI search tools such as Google AI Overviews, ChatGPT and Perplexity combine several sources into one direct answer, so a healthcare page now has to earn a citation as well as a ranking.
• A Pew Research Center analysis of 68,879 Google searches by 900 U.S. adults in March 2025 found people clicked a traditional result in 8% of visits when an AI summary appeared, against 15% when none did.
• Health content falls under Google’s Your Money or Your Life standards, so named clinical authors, last-reviewed dates and links to sources such as the CDC and NIH decide whether a page gets used.
• Question-style headings with a direct answer in the first sentence, plus bullets and tables for symptoms and steps, make healthcare pages easier for AI systems to quote.
• HIPAA risk in AI marketing sits in tools and tracking, not published content: vendors that touch protected health information need a business associate agreement, and tracking on patient portals and booking forms still needs review after the 2024 AHA v. Becerra ruling.

Patients are changing how they look for health information, and many healthcare marketing strategies haven’t caught up. Instead of scanning a list of results, more people ask full questions in ChatGPT or read Google’s AI Overviews and get an answer without visiting a website. For hospitals, practices and health systems, that shift is either a threat or an opening, depending on whether your content is built to appear in those answers.

Why does AI search behave differently than traditional search for healthcare queries?

AI search behaves differently because it writes one answer from several sources instead of showing a ranked list of pages. If someone asks Google “what are the symptoms of type 2 diabetes,” an AI Overview may answer on the results page using sources Google trusts. ChatGPT, Perplexity and similar tools work the same way. The page that used to earn the click may now earn a mention, or nothing.

The effect on clicks is measurable. A Pew Research Center analysis of 68,879 Google searches by 900 U.S. adults in March 2025 found that people clicked a traditional search result in 8% of visits when an AI summary appeared, compared with 15% when it didn’t, and clicked a link inside the summary itself in just 1% of visits.

Health questions are among the most common and most consequential searches people make. Google holds health content to its Your Money or Your Life (YMYL) standard, which means its systems are especially selective about which sources they use for medical answers. Content without clear authorship, clinical accuracy and strong sourcing is unlikely to be cited, even if it ranks well in regular search.

AI tools also change how patients ask. Because the tools are conversational, people type longer, more specific questions. “What should I expect after knee replacement surgery?” is now a common AI query, though it was rare as a typed search.

At Emulent, our view is that healthcare marketers optimizing only for traditional rankings are solving yesterday’s problem. The question now is whether your content is authoritative, well structured and specific enough to be chosen as a source when an AI system answers your patients’ most important questions.

What does E-E-A-T mean for healthcare content, and why does it matter more now?

E-E-A-T stands for Experience, Expertise, Authoritativeness and Trustworthiness, the qualities Google’s Search Quality Rater Guidelines use to judge content, with the strictest standard applied to health topics. Google’s AI Overviews draw on the same ranking systems, so content that lacks these signals is likely to be skipped as a source.

Experience means the content reflects real-world practice, such as a clinician who treats the condition or a patient educator. Expertise means the author holds the right credentials. Authoritativeness means your organization is recognized as a trusted healthcare source. Trustworthiness means your site, claims and sources can be verified.

Four ways to build E-E-A-T into healthcare content:

• Named clinical authors. Every clinical or health page should carry a byline with the author’s credentials, title and a short bio. Pages credited to “Staff Writer” carry little weight; pages by a named physician or clinical specialist carry far more. It’s one of the easiest improvements on any healthcare site.
• Publication and review dates. When a licensed clinician reviews a page, show both the publication date and a last-reviewed date with the reviewer’s name. It shows the content is current and checked, and AI systems favor content that is kept up to date.
• Links to primary sources. Cite peer-reviewed studies and clinical guidance from the CDC, the NIH and specialty societies. Content that points to reliable primary sources is more likely to be used.
• A credible About page. Your organization’s credentials, accreditations, affiliations and physical locations all add to trust. An About page that explains who you are, where you practice and what your team is qualified to do strengthens the authority of the whole site.

How should healthcare content be structured for AI-generated answers?

Healthcare content should be structured as clearly labeled sections that each answer one specific question directly. AI systems don’t read a page top to bottom; they pull out the relevant passage to build an answer. Organized, plainly written, question-led sections are cited more often than long blocks of text.

Write for the way patients phrase questions. A recovery page that buries “how long does recovery take?” in the third paragraph of a dense section is less useful to an AI system than one that opens a dedicated section with the answer and then adds detail.

Five structural elements that make healthcare content easier for AI search to use:

Structure is the part most healthcare marketing teams overlook. In our audits at Emulent, we regularly see excellent clinical content that never appears in AI answers because it’s buried in dense paragraphs with no clear organization. AI systems need to pull out a clean, accurate answer quickly; if the page isn’t built for that, it gets passed over.

How does HIPAA compliance intersect with AI-driven healthcare marketing?

HIPAA compliance intersects with AI marketing through the tools and data flows you use, not through the content you publish. Published content for AI search is public and contains no patient data. The compliance risk sits in how your organization uses AI tools internally, what you share with third-party platforms and how your website handles patient information. Any channel or tool that touches patient data, including AI chat, form submissions and ad tracking pixels, needs a review against HIPAA’s rules for protected health information (PHI).

Four HIPAA considerations for teams adopting AI-connected tools:

• Business associate agreements. Any vendor that creates, receives, stores or transmits PHI on your behalf needs a signed business associate agreement (BAA). Before adopting an AI marketing tool that touches patient or appointment data, confirm the vendor will sign one. Many general-purpose AI platforms won’t, and they shouldn’t be used in any workflow involving PHI.
• Tracking pixels and analytics. Meta and Google advertising pixels on healthcare sites have drawn enforcement actions and class-action lawsuits in recent years. In American Hospital Association v. Becerra, a federal court in Texas on June 20, 2024 vacated the part of the HHS Office for Civil Rights tracking guidance that treated an IP address plus a visit to a public, unauthenticated health page as PHI, and HHS dropped its appeal that August. The rest of the guidance still applies, so audit tracking on patient portals, appointment booking pages and any form where visitors enter health details.
• AI chat on your website. An AI chat function must be set up so it doesn’t collect or store PHI without proper safeguards. That means no pre-filling fields with patient data, clear retention limits and controlled access to transcripts.
• Staff training. Anyone who uses ChatGPT or similar tools to draft patient or marketing content should be trained never to include patient names, case details or other identifying information in a prompt. Anything typed into a third-party AI tool leaves your organization’s control.

What types of healthcare content perform best in AI search results?

The healthcare content that performs best in AI search is specific, accurate, well sourced and complete enough to answer a question without sending the reader elsewhere. Content made mainly to drive traffic, collect shares or fill a publishing calendar rarely gets cited. The test is no longer “does this rank?” but “does this answer a real patient question in a way an AI system will trust?”

Five content types that match how AI search selects healthcare sources:

• Condition and symptom explainers. Pages that cover a condition’s causes, symptoms, risk factors, diagnosis and treatment are among the most commonly cited sources in health AI answers. They must be clinically accurate, author-attributed and reviewed on a schedule.
• Procedure and treatment guides. Patients research procedures closely beforehand. Plain-language guides to what to expect, recovery time, risks and outcomes match both patient needs and how AI search works.
• Comparison and decision-support content. Pages that help patients weigh options, such as the difference between physical therapy and occupational therapy, answer high-intent questions and tend to do well in AI answers.
• Local and provider-specific pages. AI search increasingly draws on local information. Pages that state your locations, specialties and providers’ credentials help AI systems surface your organization for local patient questions.
• Patient education resources. After-visit instructions, discharge guidance and disease management guides are heavily searched after a diagnosis or procedure. Organizations that publish them openly often see them cited in AI answers for their patient populations.

How do you measure whether your healthcare content is gaining AI search visibility?

You measure AI search visibility with a mix of Search Console data, manual testing and indirect signals, because tools for tracking citations in ChatGPT or Perplexity are still immature compared with rank trackers.

Google Search Console is the most direct source today. AI Overviews appear on many health searches, and Search Console counts their impressions and clicks inside its standard Web search data rather than reporting them separately. Track impressions, clicks and click-through rate for your health topics. When a page is used in an AI Overview, impressions may stay flat or rise while clicks fall, because the answer is on the results page. That pattern isn’t necessarily a performance drop, and it fits the Pew finding above.

Four measurement approaches that give a clear picture:

• Search Console query reports. Filter for your target conditions, services and procedures and track impression and click trends over time, especially for question-style queries most likely to trigger AI Overviews.
• Manual AI query testing. Run your most important patient questions through ChatGPT, Perplexity and Google regularly and record which sources are cited. If competitors appear and you don’t, compare their structure, authorship and sourcing with yours.
• Branded search volume. When AI answers build trust in your organization, patients often follow up by searching your name. Rising branded search in Search Console is an indirect but meaningful sign of AI-driven awareness.
• Backlink and citation growth. As your content becomes a recognized source, new referring domains tend to follow. Tracking them in Ahrefs or Semrush gives a secondary signal of growing authority.

Our advice at Emulent is to start measuring AI visibility now, even imperfectly. The data you collect over the next 12 months will show what’s working long before most competitors start asking.

How do you build a healthcare marketing strategy that works across search formats?

You build a strategy that works across search formats by focusing on real patient value, accurate clinical content and visible organizational credibility. Those qualities have always earned patient trust, and they’re now the main reasons AI systems choose a source. Organizations that invest in clinical authorship, sound structure and regular review will handle the move to AI search better than those chasing volume and keywords. Some traffic will still be lost to answers on the results page; the goal is to be the name patients see in those answers.

At Emulent, we build content strategies for healthcare organizations that work in both traditional and AI search while meeting HIPAA and clinical accuracy standards, including AI search optimization. If your strategy isn’t keeping up with how patients find care, contact the Emulent team.