ChatGPT optimization for medspas

1 · What ChatGPT does when a patient asks about a medspa

ChatGPT’s default consumer model now answers local recommendation questions with web tools enabled. The patient asks “best botox in Sunnyvale”. ChatGPT issues a retrieval pass against the live web, pulls candidate citations, and composes an answer that names a small set of clinics (usually three to five) often with one or two-sentence rationales and the occasional source link.

The patient does not see the full candidate set. They see the shortlist ChatGPT chose. If your medspa is not on it, you are not in the consideration set for that query.

2 · How ChatGPT picks who to recommend

ChatGPT’s recommendation logic is not published, but the observable inputs cluster into three buckets:

• Retrieval signals. Which sources ChatGPT’s search layer surfaces for the query: major directories, local press, the clinic’s own site, third-party review aggregators.
• Authority signals. Whether those sources are well-cited elsewhere, whether the clinic’s own URL is internally consistent (name, address, services), and whether structured data confirms what the prose claims.
• Specificity signals. Whether the clinic’s presence matches the query intent: a treatment-led query rewards clinics that name the treatment explicitly on a page that can be retrieved, not just listed in a footer.

We do not have access to ChatGPT’s ranking. We do measure what shows up in its answer, and what changes when specific signals are added or fixed.

3 · Why most medspas do not appear

Common failure modes we see in audits:

• The clinic’s own site does not enumerate treatments in a way that the retrieval layer can resolve to the query.
• The clinic is absent from the third-party directories ChatGPT pulls for medspa queries in the metro.
• The clinic’s Google Business Profile, Yelp listing, and own website disagree on name, address, or services: ChatGPT downweights inconsistent entities.
• The clinic has no on-page structured data (LocalBusiness, Service, FAQPage), so ChatGPT cannot confirm what the prose claims.
• The clinic is well-known locally but has no third-party press, blog mentions, or aggregator coverage: ChatGPT has no corroborating source to cite.

4 · What we measure on ChatGPT

For each medspa, we run the 31-query basket against ChatGPT and record three quantities per query:

• Mention. Was the medspa named in ChatGPT’s answer?
• Citation. Was the medspa’s URL or property cited as a source?
• Rank position. Ordinal position among named competitors in the answer.

These roll up into a ChatGPT-specific visibility score (0–100) that sits alongside the scores for Claude, Gemini, Google AI Overviews, and Perplexity in the combined audit. Methodology detail is on the methodology page.

5 · The fix patterns that move ChatGPT rank

The audit ships the five fixes ordered by projected lift, not by activity volume. The recurring patterns:

• Treatment pages with explicit, retrievable copy: one page per major procedure, with the procedure name, location qualifier, and price band stated plainly.
• LocalBusiness, Service, and FAQPage structured data on the site, kept consistent with the prose.
• Name-address-phone consistency across Google Business Profile, Yelp, and any third-party directory ChatGPT pulls for the metro.
• Reviews that mention treatments by name: ChatGPT’s retrieval favors review corpora with semantic specificity.
• At least one third-party citation per major treatment line: a local press piece, a treatment guide, or a credible directory listing.

For the deeper how-to view written for owners doing this without an agency, see how to rank your medspa on ChatGPT.

6 · ChatGPT vs the other four surfaces

ChatGPT is one of five AI surfaces we audit. The others have different ranking logic:

• Claude (Anthropic) weights well-structured sources and reasoning chains; it is often more conservative in naming brands when sources disagree.
• Gemini (Google) sits on top of Google’s knowledge graph and Maps data, so a strong GBP and structured local-business signals move it disproportionately.
• Google AI Overviews shares some retrieval surface with Gemini but has its own citation pipeline. Treating them as one would smooth over real visibility gaps.
• Perplexity (Sonar) is a web-search answer engine that leans heavily on Reddit and other UGC for local queries, a different citation profile from the others.

A medspa can rank well on ChatGPT and poorly on Gemini, or vice versa. The audit measures all five together.

Frequently asked

• Does ChatGPT actually recommend specific medspas to patients?
  Yes. When a patient asks ChatGPT a local medspa question: "best botox in Sunnyvale", "where to get filler in 94087": ChatGPT typically names three to five specific clinics, often with a short rationale and sometimes with a citation link. Older models with web search disabled were vaguer; today’s default models with web tools enabled are direct and named.
• How often does ChatGPT update which businesses it recommends?
  Continuously. ChatGPT’s retrieval layer re-runs against the live web on most queries, so a new third-party citation, a refreshed Google Business Profile, or a new structured-data block on your site can register in days, not months. Authority shifts (which directories ChatGPT trusts for medspa data) move more slowly.
• Will paying for ads on ChatGPT help my visibility there?
  There are no medspa ad placements on ChatGPT today. ChatGPT’s recommendations are organic in the AEO sense: derived from what the surface finds and trusts about your business. There is nothing to bid on. Visibility is earned through the signals ChatGPT actually reads.
• Why does my Google Business Profile not show up in ChatGPT?
  GBP is a strong Google signal but a weak ChatGPT signal in isolation. ChatGPT cross-references multiple sources before it names a business: its own retrieval index, third-party directories (Yelp, RealSelf, local press), and citation patterns across the open web. A GBP listing without supporting citations and on-site structured data often fails to clear that bar.