Everybody's searching 'dentist near me'. Nobody can find you.

11:47pm. A mom in your zip code is holding ice to her 8-year-old's cheek. She types "emergency dentist near me" into Google. She asks ChatGPT "best pediatric dentist in [your city]". She opens Yelp. She checks Healthgrades. She finds three practices. She calls one.

That practice was not yours.

It wasn't your dentistry. Your dentistry is good. The problem is that of the four surfaces she checked in 90 seconds, your practice only appears on one of them — your own Google listing — and not at the top.

About 1.2% of local businesses are getting recommended by ChatGPT for their primary category today. For dentists, the number is roughly the same. If you're in that other 98.8%, here's the why and the fix.

The four surfaces a new patient checks

A patient looking for a dentist doesn't pick one place to search. They check three or four, fast, and trust the names that show up in more than one. Here's the typical sequence:

Google Maps "Map Pack". The three practices Google shows with map pins. Per Sistrix's CTR data, position 1 in the Map Pack gets 28.6% of clicks; position 3 drops to 11%.
ChatGPT or Perplexity. "Best dentist in [city]" or "pediatric dentist near me." BrightLocal's 2025 AI search study found AI tools cite Yelp in about 1 in 3 local queries, plus Healthgrades, Zocdoc, and local "best dentist" listicles.
Healthgrades, Zocdoc, or Vitals. Healthcare-specific directories. Patients trust them because they look clinical, not commercial. They show insurance accepted, photos, and a star rating from verified visits.
Yelp. Still the default for "is this place actually any good." Especially for cosmetic and pediatric, where decisions are emotional.

Each of these reads a different signal before deciding to show you. Being great at one is not enough.

What ChatGPT actually uses when someone asks for a dentist

When a patient types "best dentist in [your city]" into ChatGPT:

ChatGPT searches the web via Bing.
It pulls 6-15 source pages.
It picks 3-6 to cite by name.
It writes the summary the patient reads.

For dentists specifically, ChatGPT's most-cited sources are:

Healthgrades and Zocdoc profiles, weighted heavier than for most verticals because they carry insurance and procedure data
Local "best dentist" listicles from city magazines, Expertise.com, ThreeBestRated
Yelp for general sentiment
The ADA find-a-dentist directory, especially for queries like "dentist who takes [insurance]"
Your own website if it has clear service pages and FAQ schema

What ChatGPT does NOT primarily pull from: your Google Business Profile data directly. It reaches Google data indirectly through Bing, but Bing doesn't give it special weight.

Why your practice is invisible

Three patterns I see in almost every dentist audit:

1. Your GBP category is too generic

Most dentists set their primary category to "Dentist." That's the default. It's also the most competitive category in your zip code.

The win is specificity. If you do mostly pediatric work, your primary should be "Pediatric Dentist." If you do mostly Invisalign and veneers, "Cosmetic Dentist." General practice with one specialty? Primary = General, secondary = your specialty. Sterling Sky's controlled studies have shown category changes move Map Pack position more than almost any other on-profile signal.

The same logic applies to ChatGPT. When someone asks "pediatric dentist in [city]", the model preferentially pulls listings where "Pediatric Dentist" appears in the structured data, not just somewhere in the page body.

2. Your Healthgrades and Zocdoc profiles are unclaimed or empty

This is the second-biggest one. Healthgrades and Zocdoc both pre-create profiles from public data (state license records, insurance directories). Most dentists don't know theirs exist. They show up half-filled, with no photo, an old phone number, no insurance list. ChatGPT sees this and weights you down because the data is incomplete.

Fix: search your name on both sites. Claim every profile that exists. Fill in insurance accepted, procedures offered, photos, hours, languages spoken. Match your NAP (name, address, phone) to your Google listing character for character.

3. You're not on any "best dentist in [city]" lists

The Aggarwal et al. GEO paper (2024) found ChatGPT over-cites listicles by a 3:1 ratio versus long-form articles. For dentists, the listicles that matter are Expertise.com (free to apply), ThreeBestRated, your city magazine's annual "Top Dentists" issue, and Castle Connolly's regional Top Doctors list.

Apply to Expertise.com and ThreeBestRated this month. The city magazine list is usually peer-voted by other dentists, so your path in is referrals from colleagues who already made the list. Castle Connolly accepts physician/dentist nominations year-round.

What about the Map Pack?

Still drives more volume than every AI tool combined. Google search handles roughly 191 billion referrals a month versus around 1.1 billion across all AI tools (Searchable 2026). Don't ignore the Map Pack.

For dentists, four Map Pack signals carry the most weight:

Correct primary category (see above). Single biggest lever.
Review velocity. Whitespark's 2026 Local Search Ranking Factors survey rates new-review recency in the top tier. The benchmark: match your top map-pack competitor's monthly rate. For most mid-size markets, that's 6-12 new reviews per month for dentists.
Photos updated regularly. Specifically, interior photos, your team, treatment rooms, before/after when patients consent.
Service pages on your site for each procedure. "Invisalign in [city]", "dental implants in [city]", "kids dentist [neighborhood]". Each one earns its own ranking surface.

The 14 things we check at localpicks cover all of this. See the breakdown for dentists →

How to know if it's working

Pick 10 prompts a patient might actually type, and track them monthly:

"dentist near me"
"best dentist in [your city]"
"pediatric dentist [city]"
"dentist that takes [insurance] [city]"
"emergency dentist near me"
"cosmetic dentist [city]"
"Invisalign provider [city]"

Run each in ChatGPT with web search on, in Google Maps, on Healthgrades. Note which practices come up. If your name isn't there, look at who is and read their profiles. The pattern of what they share that you don't is your roadmap.

FAQ

Can I pay Healthgrades, Zocdoc, or ChatGPT to recommend me? Healthgrades and Zocdoc both offer paid placement within their own results. ChatGPT's web search results aren't pay-to-play. The free profiles on Healthgrades and Zocdoc are what ChatGPT reads, so getting those right matters more than paying for ad placement.

What about Perplexity, Gemini, Bing Copilot? Same general logic, slightly different weights. Perplexity leans on News and Reddit. Gemini favors brands with strong Wikidata presence (Andrea Volpini at WordLift has good research on this). Bing Copilot weights Bing Places and your own site. We check all of them.

How long after I fix things will I show up? Healthgrades and Zocdoc re-index within a week. Map Pack changes from a category fix usually show in 2-3 weeks. ChatGPT and Perplexity citation patterns shift in 30-60 days as their crawlers re-read updated source pages.

Do dental service pages with schema actually help? Yes. Schema.org Dentist + MedicalProcedure markup on each service page lets AI tools answer at the procedure level. Useful for queries like "who does same-day crowns in [city]" or "dentist near me that does sedation."

If you want a check on which prompts ChatGPT does and doesn't mention you for, run our free 5-minute audit. It's the same 14 checks we run on every dental practice that signs up.