“Schema Shmeema” — What ACTUALLY Drives AI Visibility for Dental Practices

In the most recent episode of the Dental Marketing Mix, Dan Brian, co-founder of DentalScapes, tackles a story that’s been circulating in dental marketing for the better part of a year — and it turns out the data doesn’t back it up.

The story goes like this: add schema markup to your website, and you’ll start showing up when patients ask ChatGPT or Google’s AI Overviews to recommend a dentist. It’s a clean pitch. It’s technical enough to sound credible, and it’s a discrete line item agencies love selling as your ticket into the age of AI search.

New research says that story is mostly wrong. Not worthless — schema still has a job to do — but wrong enough that if it’s the centerpiece of your AI visibility strategy, you’re solving the wrong problem.

The Myth: “Add Schema, Get Cited”

Schema markup is the structured data behind the scenes of your website — a labeled, machine-readable version of your practice: name, address, hours, services, reviews, insurance accepted. The logic behind treating it as an AI visibility strategy makes intuitive sense: if you hand AI systems a clean, organized data package, they should be more likely to use it.

For years, sites with rich schema also tended to rank and show up more often, so the correlation looked like causation. Dan points out that agencies — DentalScapes included, at times — have leaned on that story when talking about GEO (generative engine optimization).

What the Data Actually Shows

Ahrefs ran the most rigorous test of this claim yet. They tracked 1,885 pages that added JSON-LD schema markup between August 2025 and March 2026, matched against 4,000 similar pages that didn’t change anything, controlling for how those pages were already performing. Then they measured AI citations over the following 30 days across three platforms.

The results, using a matched difference-in-differences analysis:

• Google AI Overviews: citations went down 4.6% relative to the control group — a statistically significant decline, though a small one in absolute terms (roughly 12 fewer daily citations per page)
• Google AI Mode: +2.4%, statistically indistinguishable from zero
• ChatGPT: +2.2%, statistically indistinguishable from zero

In plain terms: adding schema markup did not increase AI citations on any of the three platforms tested. Search Engine Journal’s coverage of the study reached the same conclusion.

There’s a technical reason for this. A related experiment cited alongside the Ahrefs study found that five major AI systems — ChatGPT, Claude, Perplexity, Gemini, and Google AI Mode — extract only the visible HTML content of a page when deciding what to say about it. The structured data sitting invisibly in the code often isn’t part of that decision at all.

Where Schema Still Earns Its Keep

Dan is careful not to throw schema out entirely, and neither are the researchers behind the study. A few real reasons to keep it in place:

• It still feeds rich results, voice assistant responses, and knowledge graph entries
• It helps confirm business fundamentals — name, address, hours, entity type — for your Google Business Profile
• For a practice that isn’t showing up in AI answers at all yet, schema may still help with initial discovery — the Ahrefs study only tested pages that were already being cited regularly, so it couldn’t measure that effect

Think of schema as correctness plumbing, not a growth engine. Keep it accurate and consistent, but don’t pay a premium for it as if it’s a complete AI visibility strategy.

So What Actually Drives AI Citations?

If it’s not schema, what is? The answer isn’t a new, exotic AI-era tactic — it’s the same thing that’s driven good SEO for a decade:

• The depth and clarity of your actual, visible content
• The authority you’ve earned through genuine mentions, backlinks, and citations across the web
• Trust signals like the volume and substance of your patient reviews

AI systems are, in a real sense, reading your website the way a patient does. If your service pages are thin and your FAQ content doesn’t answer real questions in plain language, no amount of invisible code compensates for that.

This is the same idea the show covered in “Rankings to Recommendations” — AI visibility and GEO aren’t a separate discipline from SEO, they’re an extension of it. The practices winning citations in ChatGPT and AI Overviews right now are, overwhelmingly, the same ones already doing content and authority-building well.

Try it yourself: ask ChatGPT or Gemini “who’s a good dentist in [your city]?” and look at who comes up — and why. Nine times out of ten, it’s a practice with a deep, current Google Business Profile, substantive reviews, and a website that actually answers patient questions. It’s rarely, if ever, a practice that’s winning purely because a developer added Dentist schema to its homepage.

What To Do About It

• Audit your site like an AI system would. Read your service pages and FAQ content as plain text — no schema, no assumptions. Is the information a patient actually needs, in their own language, genuinely there?
• Keep your NAP (name, address, phone) identical everywhere — your website, Google Business Profile, Healthgrades, every directory. BrightLocal’s research found that 80% of consumers lose trust in a business when that information doesn’t match, and Moz’s ranking factor surveys consistently put citation consistency in the top ten for local pack visibility.
• Invest in genuine review volume and real earned authority — mentions and backlinks — instead of technical shortcuts.
• Keep your schema clean and accurate anyway. It still supports your Business Profile and rich results — just stop treating it as your AI visibility strategy.

---

TL;DR

• A rigorous Ahrefs study of 1,885 pages found that adding schema markup did not increase AI citations on Google AI Overviews, Google AI Mode, or ChatGPT — and actually correlated with a small, statistically significant decline on AI Overviews
• Most AI systems extract visible page content when generating answers, not invisible structured data
• Schema still matters for rich results and Business Profile accuracy, but it isn’t the growth lever it’s been sold as
• What actually drives AI citations is the same thing that’s always driven good SEO — genuine content depth, earned authority, and trust signals like reviews and consistent business information

---

If you want a real look at where your practice stands — not a sales pitch for schema tags, but an honest read on your content, authority, and visibility across both traditional and AI search — we’re happy to take a look.