Your Practice Website Through AI's Eyes: The Crawler & Schema Checklist for Dentists

1. Your Website Has Two Kinds of Readers Now

Patients see your website as pages: photos, colours, a friendly team picture. Search engines and AI assistants never see any of that. They send crawlers — automated programs that download your pages, extract the text and data, and file it away for later, when someone searches or asks for a recommendation. If a fact about your practice is not in what the crawler downloads, then as far as Google and the AI tools are concerned, it does not exist.

There are now two families of crawlers to care about. Google's crawler builds the index behind Google Search and Maps. Separately, the AI companies run their own crawlers to gather the information their assistants answer from — the assistants patients increasingly ask 'which dentist near me is good for implants?'. Both follow similar rules, and both can be locked out or fed empty pages by very common website setups.

Key takeaway: Crawlers read downloaded text and data, not your design. Anything they cannot download is invisible to Google and AI assistants alike.

2. Robots.txt: The Gatekeeper File

Every website has, or should have, a small text file called robots.txt: instructions addressed to crawlers, saying which parts of the site they may visit. Crawlers from reputable companies read it first and obey it. You can read yours right now — type your website address into a browser and add /robots.txt to the end.

Look for lines naming a bot and telling it to stay out. The names that matter: Googlebot is Google's crawler and should never be blocked from your public pages. For AI visibility, look for GPTBot, ClaudeBot and PerplexityBot — from OpenAI, Anthropic and Perplexity. A line pairing one of those names with 'Disallow: /' means that assistant cannot read your site at all. Blocking AI crawlers is occasionally deliberate, but for most practices it is an accident left by a template or security plugin — and it quietly removes you from AI answers while leaving your Google presence untouched, which makes it easy to miss.

Key takeaway: Read your robots.txt today. Googlebot must be allowed in, and GPTBot, ClaudeBot and PerplexityBot should be too, unless invisibility to AI assistants is a deliberate decision.

3. The JavaScript Catch: Content That Arrives Too Late

Many modern websites are built so the page arrives nearly empty and the real content is filled in afterwards by JavaScript, code that runs in the visitor's browser. Human visitors never notice. The catch is that most AI crawlers do not assemble anything: they download the page as it first arrives, and if your services, address, and phone number only appear after JavaScript runs, they take home a nearly blank page.

Google is the partial exception: its crawler can render JavaScript in a second pass — but that is slower, not guaranteed for every page, and no help at all with the AI crawlers. The test is simple: the important facts about your practice should be present in the page as first delivered, not assembled afterwards. If your site was built with a visual page-builder or app framework, this is its most likely technical problem — invisible from your own browser, which is why it survives for years.

Key takeaway: If your key content only appears after JavaScript runs, AI crawlers see a blank page. Important facts must be in the page as first delivered.

4. Schema in Plain English: Your Facts, Machine-Readable

Schema is a standard vocabulary for stating facts about your practice in a form programs cannot misread, added to a page as a small block of structured data (usually in a format called JSON-LD), hidden in the page's code where patients never see it but crawlers always read it. Instead of hoping a crawler guesses which number on your contact page is the phone number, schema states it outright: this is the phone, this is the address, these are the hours.

For a dental practice, the relevant type is called Dentist — a specific kind of local business in the schema vocabulary, which tells the crawler not just 'a business' but 'a dental practice'. The fields worth getting right are the unglamorous ones: your practice name exactly as it appears everywhere else, your full address, your phone number, your opening hours, and a field called sameAs, which lists your official profiles — your Google Business Profile, your Facebook page — so engines can connect them to the same practice.

Key takeaway: Schema states your facts explicitly instead of leaving crawlers to guess. Use the Dentist type, and get name, address, phone, hours and sameAs right.

5. Listing Your Services in Schema

Your treatments can be described in schema too. The vocabulary lets a business list what it offers — through structured offer and item-list fields, and treatment entries from the medical-procedure vocabulary — so an implants page can state, in machine-readable form, that this practice provides implant treatment — not merely mention the word in a paragraph.

Keep this in proportion. List your main services once, accurately, matching your Google Business Profile and your service pages. Never claim services you do not offer — schema that contradicts your visible pages is worse than none, because engines cross-check the two. If an agency set your site up, ask them which schema types and fields they implemented, and compare the answer against this article.

Key takeaway: List your real services in schema once, consistently with your profile and pages. Accuracy beats exhaustiveness every time.

6. Page Essentials Crawlers Depend On

Three basics decide how much of your site a crawler can use. First, crawlable service pages: one real page per main treatment, reachable by following ordinary links from your homepage. Treatments buried inside a single scrolling page, or hidden behind interactive elements a crawler cannot operate, are effectively unpublished. Second, readable text: your name, address, and phone number should appear as actual text — not only inside a logo image or a photo of your storefront sign, because crawlers cannot reliably read words out of pictures.

Third, delivery. Pages should load without errors and without making the crawler wait; a page that times out or returns an error gets skipped, and whatever was on it does not get indexed. Clear page titles help too — a title naming the treatment and the town tells the crawler exactly what the page is for. None of this needs a redesign; it is the plumbing that decides whether your website is legible to the programs that recommend you.

Key takeaway: One linkable page per service, your details as real text rather than pictures of words, and pages that load cleanly — the plumbing everything else depends on.

7. The Checklist, in Order

Run through it in this order, because each step gates the next. One: open your robots.txt and confirm Googlebot, GPTBot, ClaudeBot and PerplexityBot are not blocked. Two: confirm your key content is in the page as delivered, not assembled afterwards by JavaScript. Three: confirm Dentist schema exists and that its name, address, phone, and hours match your Google Business Profile exactly. Four: confirm each main treatment has its own crawlable page with your details in readable text.

Checking those four by hand means viewing page source and picking through code. LocoraAI's free checkup tests this layer as part of its 40 checks — crawler access, schema, service pages, and consistency — and reports what it finds, with no signup or credit card needed to see your score and your top issues. Run it, hand the results to whoever maintains your website, and the conversation stops being 'the site looks fine' and starts being a short, specific fix list.

Key takeaway: Robots access, delivered content, correct Dentist schema, crawlable service pages — and a free 40-point checkup tests all four for you.

Frequently asked questions