“Flying Blind” — Why Most Dentists Have No Idea What AI Is Saying About Them

In the most recent episode of the Dental Marketing Mix, Dan Brian, co-founder of DentalScapes, takes on a habit that almost every practice owner has picked up over the last year: opening ChatGPT, typing in “best dentist near me,” and drawing conclusions from whatever comes back. It feels like the obvious way to check your AI visibility. It’s also one of the most misleading things you can do — and practice owners are making real strategy decisions based on it.

What You See Is Not What Your Patients See

Dan’s core point is simple: AI search isn’t like Googling something in 2012. When a patient asks ChatGPT, Google’s AI Mode, or Gemini “who’s the best dentist near me for Invisalign,” the platform isn’t pulling a neutral, universal answer off a shelf. It’s generating a response shaped by an enormous number of variables:

• Which platform and device they’re using
• Their location and whether they’re logged in
• Their search history and the queries they’ve made before
• Which links they’ve previously clicked

In the case of Gemini and AI Overviews specifically, Google has already confirmed it actively uses personal search history to personalize and deepen AI responses. As Dan puts it, it’s not a search engine anymore — it’s a personal AI assistant that increasingly knows who you are and what you care about.

Which is exactly why your own results are unreliable. When you type “best dentist in Miami” into ChatGPT, you’re seeing a response shaped by the fact that you are a dentist, that you’ve searched your own practice name dozens of times, that you’ve clicked your own website, and that the platform has profiled you as someone deeply connected to that practice. The answer you get back is almost certainly more favorable — or at minimum completely different — than what a first-time patient searching cold would see.

Dan’s analogy lands: it’s like checking your Yelp reviews while logged into your own account and wondering why everyone seems to love you.

There Is No Single “AI Search Result” to Check

There’s a second layer to this, and it’s more technical but just as important: these platforms don’t work the same way as each other.

• ChatGPT pulls heavily from third-party directories and listings — Yelp, Healthgrades, and similar sources
• Gemini leans more on brand-owned websites and structured data
• Perplexity and Claude each have their own sourcing logic

What shows up for your practice on one platform may look nothing like what shows up on another — and a patient might be using any one of them. There’s no equivalent of checking a single rank position the way there was with traditional Google results. There are dozens of possible outputs depending on the platform, the phrasing of the query, the user’s context, and the moment in time.

That last variable matters more than people realize. These systems aren’t static. Models are being retrained, new sources are being indexed, and the platforms are evolving constantly. The answer you get today may not be the answer a patient gets next week.

An Honest Look at Where AI Search Measurement Stands

Dan is deliberately candid in this section of the episode: AI search measurement is genuinely new territory. The tools that exist to track AI visibility are early stage. They’re improving fast, but none of them deliver the clean, reliable data you’d get from traditional SEO rank tracking.

There’s no AI equivalent of checking where you rank for “dentist in Detroit” and seeing a clear number. The outputs are generative, variable, and highly contextual. That’s simply where the industry is right now, and Dan would rather say so than oversell it.

What the best measurement approaches can do today is simulate queries across platforms in a consistent, depersonalized way — asking cold, from outside your own account, across a range of relevant queries, and tracking what comes back over time. That gives a far more honest picture than running searches yourself, and it can tell you:

• Whether your practice is being mentioned at all
• Whether competitors are showing up when you’re not
• Whether AI platforms appear to trust your website as a source
• Whether your directory and citation presence is strong enough to get picked up by platforms that rely on third-party sources

That’s useful intelligence. A screenshot you took on your phone after Googling yourself is not.

Treat AI Visibility Like a Real Channel — Even While Measurement Is Imperfect

The dentists who win in AI search, Dan argues, aren’t the ones who occasionally check their own results and feel good or bad about what they see. They’re the ones treating AI visibility as a real, measurable channel and building a strategy to improve it systematically. In practice, that means four things:

1. Make Sure AI Platforms Can Actually Parse Your Website

Structure matters. If your services, locations, and differentiators aren’t clearly stated in text these systems can read, they have nothing to work with when a patient asks a question you should be the answer to.

2. Strengthen Your Directory and Citation Presence

Platforms like ChatGPT lean heavily on third-party listings. If your presence across directories is thin or inconsistent, you’re invisible on the platforms that source that way — no matter how good your website is.

3. Understand the Questions Patients Are Actually Asking

AI queries are longer, more specific, and more personal than the keywords practices are used to targeting. Knowing what patients are asking — and whether your practice is positioned to appear in those answers — is the strategic work.

4. Track Visibility Over Time, Consistently and Depersonalized

One-off checks tell you nothing. A consistent, cold, cross-platform baseline tracked over time tells you whether you’re gaining ground.

Build the Foundation Before the Tools Catch Up

This is the work DentalScapes is doing with clients right now. Rather than waiting for measurement tools to mature, the focus is on building the foundation — website structure, content, the citation ecosystem, and authority signals — that positions practices to show up as AI search matures and the measurement picture gets clearer.

The practices that build that foundation now will have a real advantage over the ones who wait until everything is figured out.

---

TL;DR

• Checking your own AI results is fundamentally flawed — platforms personalize responses using your history, clicks, location, and profile
• Google has confirmed Gemini and AI Overviews use personal search history to personalize answers
• As a dentist searching your own practice, you almost certainly see a more favorable or entirely different answer than a cold, first-time patient
• Platforms source differently: ChatGPT leans on third-party directories, Gemini on brand-owned sites and structured data, Perplexity and Claude on their own logic
• There is no single AI search result to check — outputs vary by platform, phrasing, user context, and moment in time
• AI search measurement is early stage; the honest approach is simulating queries cold and depersonalized across platforms and tracking trends over time
• Good measurement can reveal whether you’re mentioned at all, whether competitors appear when you don’t, and whether platforms trust your site and citations
• The winning move now: make your site parseable, strengthen citations, know the questions patients ask AI, and track visibility consistently

---