One lets AI in. One built the better map.

A patient types dermatologist near me that takes my insurance. Two big, multi-state dermatology groups want that visit: Schweiger Dermatology, with 158 location pages, and Forefront Dermatology, with more than 280 locations by its own count.

To recommend one, an AI assistant has to connect five things: the condition, the insurance, the clinic, the right specialist and a way to book. The easier a website makes those connections for a machine, the easier it is to recommend.

What Schweiger does right: every clinic page opens

We read 30 Schweiger clinic pages at random on 14 October 2026, before JavaScript runs. A browser and Googlebot got the same page. Every one carried a business listing with the street, phone and opening hours, linked to that clinic’s own providers, and offered a booking link.

Schweiger also wrote a note for the AI engines: a 6 KB llms.txt with 46 links, pointing assistants at its insurance list, its locations by state, and its Mohs and allergy specialists.

Where Schweiger stops: the connections live in the words

A reader sees the providers, the treatments and, on about half the pages, the insurers. The listing a machine reads carries the address and the hours, and stops there.

"@type": "LocalBusiness", "telephone", "address",
"openingHoursSpecification": [ … ]

No map pin. No rating. No treatment. No provider. No booking action.

One thing to fix this week: two sets of opening hours

Next to each clinic’s real hours, every Schweiger clinic page also carries a group-wide entry for the head office in New York that says it is open seven days a week, 9 to 5. An assistant asked “is it open on Sunday?” can read either one.

LocalBusiness, 400 West Main Street → Sat 07:00–14:00, Sunday closed
Organization, 33 East 33rd Street (corporate) →
"Monday,Tuesday,Wednesday,Thursday,Friday,Saturday,Sunday 09:00-17:00"

All 30 clinic pages we sampled carried both.

Forefront: we could not open it with any automated reader

Every route we tried on 14 October 2026 got the same “Just a moment…” challenge page, while Google still lists its clinic pages. The door opens for some machines and not others.

plain request, home page          → 403 challenge
GPTBot · ClaudeBot · PerplexityBot → 403 challenge
headless browser                   → 403 challenge
robots.txt · llms.txt · sitemap    → 403 challenge
Google search                      → clinic pages indexed

So we make no claims about what Forefront’s pages contain in their code. We are reporting what an ordinary machine gets when it knocks: a page asking it to prove it is human.

What a person sees: the better map

Through Google’s index, Forefront’s pages show exactly the connections this search needs: a location finder with an insurance dropdown, an insurance list on each clinic page, a team page per clinic you can filter by service, and specialists by type, from Mohs surgeons to cosmetic experts.

The pattern across this series: each brand built the half the other skipped. Schweiger opened the door. Forefront drew the map.

Why this matters to a clinic

The easier your site makes the connections, the easier AI recommends you.

1. Name the conditions you treat and the insurers you take, in the page data and not only in the words.
2. Attach your providers to each clinic, so an assistant can match the right specialist to the right location.
3. Give each clinic one set of opening hours.
4. Make sure the AI assistants can get through the door.

• We search for your clinic the way a patient would. Your city and your competitors, on Google and on the AI assistants.
• You get the gaps in plain sentences. One list you can hand to whoever runs your website.
• It takes a day. No commitment, no cost.

Download the slides (PDF, 1.2 MB)