Med spas get recommended by ChatGPT and Google's AI answers when the sources those engines already trust - a complete Google Business Profile, review platforms, directories, and third-party lists - describe the practice consistently and specifically. Organic rank does not carry over. The two systems retrieve from different places, and one of them barely reads your homepage.
Ranking first does not put you in the answer
Rank and citation are decided separately. Google's AI features draw on its Search index but assemble answers from passages, so a page sitting third can be quoted while the page at the top is not. ChatGPT and Perplexity run their own crawlers and their own retrieval, and neither one consults your Google position at all.
Google's own developer documentation on AI features is blunt about the mechanics: there is no special markup, no submission process, and no separate opt-in that makes a page eligible for an AI answer. Eligibility follows ordinary Search eligibility. Anyone selling a med spa a distinct GEO product with proprietary AI schema is selling something Google says does not exist.
What does differ is the unit of retrieval. Search returns pages; an AI answer stitches together sentences it can verify across several sources. A claim that appears only on your own website is a claim with one witness. The same claim echoed on your Google Business Profile, in your directory listings, and in a third-party roundup is a fact the model can state without hedging.
Med spas get asked two very different questions
Aesthetic queries split cleanly. Treatment questions - is Botox safe, how long does filler last - get answered from medical authorities. Provider questions - who should I see in Charlotte for lip filler - get answered from local sources. A med spa can realistically be named in the second category and almost never in the first.
That split is not arbitrary. Health sits inside what Google's published Search Quality Rater Guidelines call Your Money or Your Life topics, where the guidelines instruct raters to hold pages to a higher standard of expertise and trust. Those guidelines describe how Google evaluates quality broadly, and the practical effect on AI answers is visible: safety and efficacy questions get sourced to institutions, regulators, and peer-reviewed literature, not to practices selling the treatment.
Most med spa content strategies get this exactly backwards. They produce a blog full of general treatment explainers competing against the FDA and academic dermatology, and publish almost nothing about the specific practice - who works there, what a visit involves, how pricing is structured, which conditions are treated and which are referred out. The explainers cannot win. The practice-specific pages have essentially no competition.
What actually gets cited for "best med spa near me"
For provider-choice queries, AI answers are assembled mostly from map and directory data, review platforms, and best-of roundups - sources that describe a practice without selling for it. Your own website carries one category almost entirely, which is whether you perform a given treatment, and that is the category most practices document worst.
| The question asked | What the engines lean on | What you control |
|---|---|---|
| Is this treatment safe / how does it work | Regulators, hospital systems, specialty academies, published studies | Very little - do not build a strategy here |
| Best med spa in [city] / who should I see | Map and directory data, review platforms, best-of roundups | Profile completeness, review volume and recency, being listed at all |
| Does [practice] do [treatment] | Your own site and your Google Business Profile services list | Nearly all of it - and most spas leave it half filled in |
| What does [treatment] cost in [city] | Practice pages that state pricing mechanics, aggregator estimates | Whether your page explains per-unit vs per-area pricing at all |
| Who is qualified to perform this in [state] | State medical and nursing board material, statutes | Whether your provider credentials and medical director are named |
First, check that you are not blocking the crawlers
A site that blocks AI crawlers cannot be cited by them, and plenty of med spas block without knowing it. Security plugins, bot rules, and hosting defaults do it quietly. Cloudflare announced in July 2025 that new domains on its network default to blocking AI crawlers - sound policy for a publisher, usually wrong for a practice that wants to be recommended.
The distinction that matters is training versus retrieval. Blocking a training crawler keeps your content out of a future model. Blocking a search or user-triggered fetcher removes you from the answer being generated right now, for a person who is actively looking for a provider. Those are different decisions and most robots.txt files treat them as one.
| Agent | What it does | Cost of blocking it |
|---|---|---|
| OAI-SearchBot | Indexes pages for search results shown inside ChatGPT | You stop appearing in ChatGPT's search answers |
| ChatGPT-User | Fetches a page because a user's prompt asked for it | ChatGPT cannot read your page even when asked to |
| GPTBot | Collects content OpenAI may use for model training | Training only - blocking is a defensible business choice |
| Google-Extended | Controls use of content for Gemini and grounding, per Google's docs | Google states it does not affect Search inclusion or AI Overviews |
| PerplexityBot | Indexes pages so Perplexity can cite and link them | You lose the citation surface Perplexity is built around |
| ClaudeBot | Anthropic's web crawler, documented alongside its user-triggered fetchers | Your pages stop being available to Claude's retrieval |
Read the operators' own bot documentation before editing anything - OpenAI, Google, Perplexity, and Anthropic each publish a current list of user agents and IP ranges, and the names change. Then confirm your robots.txt matches what you decided rather than what a plugin decided in 2023.
The pages that get quoted are specific and boring
A passage gets pulled into an AI answer when it states something concrete that the model cannot get from three other sources. For a med spa that means the operational detail nobody else can publish about you: the exact treatments performed, by whom, with what credential, how pricing is structured, what the appointment involves, and who is not a candidate.
Write in answer-first order. Put the direct answer in the first two or three sentences under each heading and elaborate afterward - retrieval works on passages, and a passage that buries its answer under a paragraph of scene-setting is a passage that does not get selected. Our med spa SEO guide covers the ranking side of the same pages.
Resist the pull toward volume. Google's spam policies name scaled content abuse - producing pages at volume primarily to manipulate rankings rather than to help people - as a distinct violation, and the policy applies regardless of whether the pages were written by a person or a machine. Thirty thin treatment explainers is the exact pattern that policy describes. Six pages that say something only your practice knows is the opposite bet, and it is the one that gets cited.
Your off-site record does most of the work
For provider-choice questions, the deciding evidence sits outside your website. Directory listings, review platforms, and best-of articles are what the engines read when someone asks who to see, because those sources describe you without a commercial interest in the answer. This is the part med spas most often neglect while redesigning a homepage.
Three things carry disproportionate weight. Consistency first: an identical practice name, address, phone number, and description everywhere, because a model resolving contradictory records tends to state nothing rather than guess. Reviews second - volume and recency both, replied to compliantly, since a public reply confirming someone was a patient is its own privacy problem. Third, a services list that is actually filled in; the Google Business Profile services section is a structured statement of what you treat, and half-empty profiles are the norm, which is covered field by field in our profile setup guide.
Being named in a third-party comparison is worth more than another page on your own site. That is not a shortcut to buy - paid placement in a low-quality directory is a link scheme by Google's definition - but it is a reason to respond when a legitimate local publication or roundup asks about your practice.
Health topics raise the bar you have to clear
Chasing AI citations suspends none of the rules that already govern med spa marketing. Results claims still fall under the FTC's Endorsement Guides at 16 CFR Part 255, patient photos and review replies still carry privacy obligations, and in North Carolina who may perform and delegate cosmetic procedures is governed by the NC Medical Board and the Board of Nursing.
The overlap works in your favor more often than not. Naming your providers with their credentials and identifying the supervising physician is a compliance posture and simultaneously the strongest expertise signal on the site. Stating candidacy limits honestly is both accurate and the kind of specificity that gets quoted. Where the two diverge - an unverifiable outcome claim that would read well in an AI summary - compliance wins, and the claim comes out.
A visibility check you can run this week
Auditing AI visibility takes an afternoon and no software. Ask the engines the questions your patients ask, record who gets named and what gets cited, then confirm you are reachable and your listings agree with each other. The cited sources are the work list; everything else is guessing at a black box.
- 1Ask ChatGPT, Perplexity, and Google the five questions your patients actually ask, with your city attached. Write down who gets named and which sources get cited.
- 2That list of cited sources is your work list - if a directory or review platform keeps appearing and you are not on it, that is the gap.
- 3Open yoursite.com/robots.txt and confirm you are not blocking OAI-SearchBot, ChatGPT-User, or PerplexityBot by accident.
- 4Check your host or CDN separately - a bot-blocking default can stop a crawler that robots.txt welcomed.
- 5Ask an engine directly what it knows about your practice. Wrong hours, a stale address, or a treatment you no longer offer tells you which listing is feeding it.
- 6Confirm your Google Business Profile services section lists every treatment you actively market, in the words patients use.
- 7Pick your three highest-value treatment pages and rewrite the opening of each section to answer its heading in the first two sentences.
Sources
- Google Search Central, AI features and your website - States there are no additional requirements, markup, or structured data needed to appear in AI Overviews or AI Mode.
- Google Search Central, Spam policies for Google web search (scaled content abuse) - Defines scaled content abuse as pages generated primarily to manipulate rankings, no matter how they are created.
- Google, Search Quality Rater Guidelines - General Guidelines (September 11, 2025), section 2.3 Your Money or Your Life topics
- Google Search Central, List of Google's common crawlers (Google-Extended) - Google-Extended controls training and grounding use and does not impact a site's inclusion in Google Search.
- OpenAI, Overview of OpenAI crawlers (OAI-SearchBot, ChatGPT-User, GPTBot)
- Perplexity, Perplexity Crawlers (PerplexityBot)
- Anthropic, Does Anthropic crawl data from the web, and how can site owners block the crawler? (ClaudeBot, Claude-User, Claude-SearchBot)
- Cloudflare, Content Independence Day: no AI crawl without compensation (July 1, 2025) - The announcement that changed Cloudflare's default for AI crawlers.
- 16 CFR Part 255, Guides Concerning the Use of Endorsements and Testimonials in Advertising (Cornell LII)
- North Carolina Medical Board, Position Statements (5.1.2 Laser Surgery; 9.1.1 Physician Supervision of Other Licensed Health Care Professionals)