Canonical: https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study

Rotgar Research · Aesthetic clinics · United States

# Aesthetic Clinic AI Search Visibility Across Four Platforms

A US aesthetic-clinic query panel comparing answers and cited source sets across ChatGPT, Gemini, Google AI Overviews and Google AI Mode.

By [**Evgeniy Yudin**](https://rotgar.com/about), Founder and Strategy Lead · Published October 27, 2026 · Version 1.0 · US English

[Read the findings](https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study#answer) [Review the methodology](https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study#methodology)

Direct answer

## Track the aesthetic procedure and the platform separately

We tested 18 fixed US-English aesthetic-clinic questions across four platforms in three same-day waves. The panel covers neuromodulator injections, dermal fillers and laser skin resurfacing in New York City, Chicago and Miami.

ChatGPT: 54/54 checks contained an AI answer; Gemini: 54/54 checks contained an AI answer; Google AI Overviews: 43/54 checks contained an AI answer; Google AI Mode: 54/54 checks contained an AI answer.

Among 205 answer-bearing captures, 182 contained at least one eligible citation. The tables separate answer availability, cited-source overlap and repeatability; these are different observations, not interchangeable visibility scores.

**216**

### Planned core checks

18 query cells × 3 waves × 4 platforms. Technical preflight is excluded.

**18**

### Exact query cells

3 cities × 3 nonsurgical procedures × 2 wording variants.

**180**

### Defined repeat-source pairs

Hostname Jaccard has a nonempty citation union; empty/empty sets are not counted as 100% stability.

Study at a glance

## A fixed aesthetic clinic panel, separated by platform

Patient tasks

Neuromodulator injections, dermal fillers and laser resurfacing

Requested US markets

New York City, Chicago and Miami

Collection surfaces

ChatGPT, Gemini, Google AI Overviews and Google AI Mode through DataForSEO

Repeated collection

Three same-day waves of the same 18 query cells

Fieldwork

October 27, 2026

Public evidence

Versioned data, prompts, pair tables, source ledger and reproduction materials

What the panel shows

## Findings for aesthetic clinic teams

### The same tracking sheet needs separate platform columns

ChatGPT: 54/54 checks contained an AI answer; Gemini: 54/54 checks contained an AI answer; Google AI Overviews: 43/54 checks contained an AI answer; Google AI Mode: 54/54 checks contained an AI answer.

Record whether an answer exists before assessing a named clinic or practitioner. A missing AI Overview is not a clinic losing a recommendation.

### A repeated question is a baseline—not an implementation result

Mean within-query source-hostname Jaccard in this panel was ChatGPT 24.6% (54 defined pairs); Gemini 7.8% (38 defined pairs); Google AI Overviews 23.9% (34 defined pairs); Google AI Mode 15.2% (54 defined pairs).

Keep the exact wording, city, platform and collection date alongside each check. Compare repeated baseline observations before attributing a change to website work.

### Inspect the cited page, not just the practice domain

1259 deduplicated URL occurrences were visible across the core answers. Treatment / services path was the largest conservative URL category (467 occurrences).

The URL-type table is a way to find procedure, practitioner and location pages worth inspecting. It does not show that adding a particular page caused an AI recommendation.

### Neuromodulator, filler and resurfacing questions have distinct scopes

The procedure table keeps neuromodulator injections, dermal fillers and laser skin resurfacing separate in each system. The two wordings differ: one asks for clinics to compare; the other explicitly asks for the practitioner, clinic location and source evidence.

Use these segments to select procedure pages, named-practitioner profiles and branch-specific information for inspection. No automated clinic-accuracy or clinical-quality score is inferred.

Four platforms

## What each platform returned for the fixed aesthetic-clinic panel

All 216 planned observation slots are retained. Captured outcomes separate an AI answer from a successfully completed check with no AI answer. Any gap between planned and captured slots is a technical collection state, documented in the methodology and observation download—not clinic-brand absence.

**What each platform returned for the fixed aesthetic-clinic panel · Frozen public v1.0 panel**

| Platform | Planned slots | Captured | Answer-bearing | No AI answer | With citations | Unique URLs | Unique hostnames |
| --- | --- | --- | --- | --- | --- | --- | --- |
| ChatGPT | 54 | 54 | 54 | 0 | 54 | 346 | 199 |
| Gemini | 54 | 54 | 54 | 0 | 31 | 104 | 81 |
| Google AI Overviews | 54 | 54 | 43 | 11 | 43 | 247 | 123 |
| Google AI Mode | 54 | 54 | 54 | 0 | 54 | 197 | 101 |

Treatment and task

## Keep nonsurgical procedures separate

Each platform–task segment contains 18 planned checks: three cities, two wording variants and three scheduled repeats. Captured outcomes and answer-bearing captures have separate counts. No segment is weighted by patient traffic or query volume.

**Keep nonsurgical procedures separate · Frozen public v1.0 panel**

| Platform | Procedure | Planned slots | Captured | Answer-bearing | No AI answer | With citations | Unique hostnames |
| --- | --- | --- | --- | --- | --- | --- | --- |
| ChatGPT | Neuromodulator injections | 18 | 18 | 18 | 0 | 18 | 89 |
| ChatGPT | Dermal fillers | 18 | 18 | 18 | 0 | 18 | 83 |
| ChatGPT | Laser skin resurfacing | 18 | 18 | 18 | 0 | 18 | 76 |
| Gemini | Neuromodulator injections | 18 | 18 | 18 | 0 | 10 | 37 |
| Gemini | Dermal fillers | 18 | 18 | 18 | 0 | 11 | 34 |
| Gemini | Laser skin resurfacing | 18 | 18 | 18 | 0 | 10 | 29 |
| Google AI Overviews | Neuromodulator injections | 18 | 18 | 14 | 4 | 14 | 54 |
| Google AI Overviews | Dermal fillers | 18 | 18 | 17 | 1 | 17 | 49 |
| Google AI Overviews | Laser skin resurfacing | 18 | 18 | 12 | 6 | 12 | 53 |
| Google AI Mode | Neuromodulator injections | 18 | 18 | 18 | 0 | 18 | 46 |
| Google AI Mode | Dermal fillers | 18 | 18 | 18 | 0 | 18 | 45 |
| Google AI Mode | Laser skin resurfacing | 18 | 18 | 18 | 0 | 18 | 39 |

Repeated questions

## How much did cited sources repeat within the same question?

Three pairwise comparisons are planned for each exact query–platform cell. Means use answer-bearing pairs with a nonempty source union. One-empty-source pairs count as zero; empty/empty pairs are undefined. The both-nonempty column is a separate sensitivity view.

### Visible citation sets when the same aesthetic clinic question was repeated

Mean within-cell pairwise Jaccard. A larger value means more of the same cited sources, not better answers.

Exact URLs Source hostnames

#### ChatGPT

54 defined hostname pairs from 54 planned pairs; 0 source-empty/source-empty answer pairs.

#### Gemini

38 defined hostname pairs from 54 planned pairs; 16 source-empty/source-empty answer pairs.

#### Google AI Overviews

34 defined hostname pairs from 54 planned pairs; 0 source-empty/source-empty answer pairs.

#### Google AI Mode

54 defined hostname pairs from 54 planned pairs; 0 source-empty/source-empty answer pairs.

**How much did cited sources repeat within the same question? · Frozen public v1.0 panel**

| Platform | Planned pairs | Answer pairs | Defined host pairs | Both source-empty | One source-empty | URL Jaccard | Hostname Jaccard | Both-nonempty host Jaccard |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| ChatGPT | 54 | 54 | 54 | 0 | 0 | 18.0% | 24.6% | 24.6% |
| Gemini | 54 | 54 | 38 | 16 | 14 | 5.6% | 7.8% | 12.4% |
| Google AI Overviews | 54 | 34 | 34 | 0 | 0 | 15.9% | 23.9% | 23.9% |
| Google AI Mode | 54 | 54 | 54 | 0 | 0 | 8.5% | 15.2% | 15.2% |

Matched source sets

## How similar were source sets across platforms?

Matched pairs share the exact query and scheduled repeat label. Each platform combination has 54 planned matched pairs. Actual timestamps—including delayed recoveries—are retained. Only answer-bearing captures enter source-set comparisons; no-answer and technical-error pairs are not treated as empty source sets.

**How similar were source sets across platforms? · Frozen public v1.0 panel**

| Platform pair | Planned matched pairs | Answer pairs | Defined pairs | Both source-empty | Hostname Jaccard | Both-nonempty Jaccard |
| --- | --- | --- | --- | --- | --- | --- |
| ChatGPT / Gemini | 54 | 54 | 54 | 0 | 2.2% | 3.8% |
| ChatGPT / Google AI Overviews | 54 | 43 | 43 | 0 | 3.2% | 3.2% |
| ChatGPT / Google AI Mode | 54 | 54 | 54 | 0 | 3.2% | 3.2% |
| Gemini / Google AI Overviews | 54 | 43 | 43 | 0 | 1.7% | 3.1% |
| Gemini / Google AI Mode | 54 | 54 | 54 | 0 | 2.5% | 4.3% |
| Google AI Overviews / Google AI Mode | 54 | 43 | 43 | 0 | 11.6% | 11.6% |

Source-page structure

## What kinds of citation URLs were visible?

Counts are unique URL occurrences within an answer, then summed across answers. The same URL can recur in different answers. Categories use conservative hostname/path rules, not a content-quality or ranking-factor assessment.

**What kinds of citation URLs were visible? · Frozen public v1.0 panel**

| Platform | URL category | Citation occurrences | All citation occurrences | Share |
| --- | --- | --- | --- | --- |
| ChatGPT | Homepage | 139 | 536 | 25.9% |
| ChatGPT | Treatment / services path | 198 | 536 | 36.9% |
| ChatGPT | Practitioner / team path | 11 | 536 | 2.1% |
| ChatGPT | Location / contact path | 29 | 536 | 5.4% |
| ChatGPT | Editorial path | 1 | 536 | 0.2% |
| ChatGPT | Known directory / social host | 33 | 536 | 6.2% |
| ChatGPT | Unclassified path | 125 | 536 | 23.3% |
| Gemini | Homepage | 15 | 133 | 11.3% |
| Gemini | Treatment / services path | 66 | 133 | 49.6% |
| Gemini | Practitioner / team path | 0 | 133 | 0.0% |
| Gemini | Location / contact path | 9 | 133 | 6.8% |
| Gemini | Editorial path | 6 | 133 | 4.5% |
| Gemini | Known directory / social host | 2 | 133 | 1.5% |
| Gemini | Unclassified path | 35 | 133 | 26.3% |
| Google AI Overviews | Homepage | 16 | 334 | 4.8% |
| Google AI Overviews | Treatment / services path | 112 | 334 | 33.5% |
| Google AI Overviews | Practitioner / team path | 3 | 334 | 0.9% |
| Google AI Overviews | Location / contact path | 8 | 334 | 2.4% |
| Google AI Overviews | Editorial path | 5 | 334 | 1.5% |
| Google AI Overviews | Known directory / social host | 26 | 334 | 7.8% |
| Google AI Overviews | Unclassified path | 164 | 334 | 49.1% |
| Google AI Mode | Homepage | 28 | 256 | 10.9% |
| Google AI Mode | Treatment / services path | 91 | 256 | 35.5% |
| Google AI Mode | Practitioner / team path | 4 | 256 | 1.6% |
| Google AI Mode | Location / contact path | 9 | 256 | 3.5% |
| Google AI Mode | Editorial path | 0 | 256 | 0.0% |
| Google AI Mode | Known directory / social host | 12 | 256 | 4.7% |
| Google AI Mode | Unclassified path | 112 | 256 | 43.8% |

Selected source-page follow-up

## A predetermined source-page follow-up

The first declared citation in the selection-wording query of scheduled wave 1 supplies one slot per city × procedure × platform: 36 slots. A delayed same-input recovery can fill its original failed slot and is flagged in the download. No alternate wording or more convenient citation replaces the selected source.

**A predetermined source-page follow-up · Frozen public v1.0 panel**

| Slot | Platform | City | Procedure | Selected source | Review state | Treatment detail | Practitioner detail | Location detail | Contact detail | Fee / payment context |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Q01 | ChatGPT | New York City | Neuromodulator injections | [dorisdaymd.com/pages/neuromodulators](https://dorisdaymd.com/pages/neuromodulators) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q02 | Gemini | New York City | Neuromodulator injections | [medspafind.com/blog/best-med-spas-in-new-york-city-2026](https://medspafind.com/blog/best-med-spas-in-new-york-city-2026) | Inspected static page | Present | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Present |
| Q03 | Google AI Overviews | New York City | Neuromodulator injections | [getplump.com/blog/botox-cost-nyc-2026](https://getplump.com/blog/botox-cost-nyc-2026) | Inspected static page | Present | Unknown / not assessed | Unknown / not assessed | Present | Present |
| Q04 | Google AI Mode | New York City | Neuromodulator injections | [m.yelp.com/search](https://m.yelp.com/search?find_desc=Botox+And+Fillers&find_loc=New+York%2C+NY) | Page request blocked | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q05 | ChatGPT | New York City | Dermal fillers | [dorisdaymd.com/pages/fillers](https://dorisdaymd.com/pages/fillers) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q06 | Gemini | New York City | Dermal fillers | [icreativez.com/aesthetics-clinics-local-map-city/best-aesthetics-clinics-new-york-city-2026.html](https://icreativez.com/aesthetics-clinics-local-map-city/best-aesthetics-clinics-new-york-city-2026.html) | Inspected static page | Present | Present | Unknown / not assessed | Unknown / not assessed | Present |
| Q07 | Google AI Overviews | New York City | Dermal fillers | [drarielostad.com/combined-surgical-non-surgical-facial-rejuvenation-nyc](https://drarielostad.com/combined-surgical-non-surgical-facial-rejuvenation-nyc) | Inspected static page | Unknown / not assessed | Present | Present | Present | Unknown / not assessed |
| Q08 | Google AI Mode | New York City | Dermal fillers | [nextpatient.co/p/3178/providers](https://nextpatient.co/p/3178/providers) | uninspectable\_booking\_static\_shell | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q09 | ChatGPT | New York City | Laser skin resurfacing | [mountsinai.org/care/ent/services/facial-plastic-reconstructive-surgery/laser-resurfacing](https://mountsinai.org/care/ent/services/facial-plastic-reconstructive-surgery/laser-resurfacing) | Inspected static page | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Present | Unknown / not assessed |
| Q10 | Gemini | New York City | Laser skin resurfacing | [dermatologyandlasergroup.com](https://dermatologyandlasergroup.com/) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q11 | Google AI Overviews | New York City | Laser skin resurfacing | — | No selected source: no AI answer | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q12 | Google AI Mode | New York City | Laser skin resurfacing | [nextpatient.co/p/4588/providers](https://nextpatient.co/p/4588/providers) | uninspectable\_booking\_static\_shell | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q13 | ChatGPT | Chicago | Neuromodulator injections | [chicagodermatology.com](https://chicagodermatology.com/) | Inspected static page | Unknown / not assessed | Present | Unknown / not assessed | Present | Unknown / not assessed |
| Q14 | Gemini | Chicago | Neuromodulator injections | [chicagoaesthetics.com/med-spa-treatments/dysport](https://chicagoaesthetics.com/med-spa-treatments/dysport) | Inspected static page | Present | Present | Present | Present | Present |
| Q15 | Google AI Overviews | Chicago | Neuromodulator injections | — | No selected source: no AI answer | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q16 | Google AI Mode | Chicago | Neuromodulator injections | [m.yelp.com/search](https://m.yelp.com/search?find_desc=Botox&find_loc=Chicago%2C+IL) | Page request blocked | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q17 | ChatGPT | Chicago | Dermal fillers | [google.com/maps/place](https://google.com/maps/place?q=place_id%3AChIJ9f8w5LEsDogRlNwDp1kHXnU&hl=en) | Navigation / booking surface | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q18 | Gemini | Chicago | Dermal fillers | [imagelabmedspa.com/locations/chicago](https://imagelabmedspa.com/locations/chicago) | Inspected static page | Present | Present | Present | Present | Present |
| Q19 | Google AI Overviews | Chicago | Dermal fillers | [marcoellismd.com/facial-surgery/non-surgical-facial-rejuvenation](https://marcoellismd.com/facial-surgery/non-surgical-facial-rejuvenation) | Inspected static page | Present | Present | Present | Present | Present |
| Q20 | Google AI Mode | Chicago | Dermal fillers | [m.yelp.com/search](https://m.yelp.com/search?find_desc=Medical+Spas&find_loc=Chicago%2C+IL) | Page request blocked | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q21 | ChatGPT | Chicago | Laser skin resurfacing | [ericksondermatology.com/services/lasers](https://ericksondermatology.com/services/lasers) | Inspected static page | Present | Present | Present | Present | Present |
| Q22 | Gemini | Chicago | Laser skin resurfacing | — | not\_applicable\_answer\_without\_sources | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q23 | Google AI Overviews | Chicago | Laser skin resurfacing | — | No selected source: no AI answer | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q24 | Google AI Mode | Chicago | Laser skin resurfacing | [nextpatient.co/p/6421/providers](https://nextpatient.co/p/6421/providers) | uninspectable\_booking\_static\_shell | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q25 | ChatGPT | Miami | Neuromodulator injections | [med.miami.edu/departments/dermatology/divisions-and-programs/cosmetic-dermatology](https://med.miami.edu/departments/dermatology/divisions-and-programs/cosmetic-dermatology) | Inspected static page | Present | Present | Unknown / not assessed | Present | Unknown / not assessed |
| Q26 | Gemini | Miami | Neuromodulator injections | [monacomedspa.com](https://monacomedspa.com/) | Inspected static page | Present | Present | Present | Present | Present |
| Q27 | Google AI Overviews | Miami | Neuromodulator injections | [m.yelp.com/search](https://m.yelp.com/search?find_desc=Botox+Injections&find_loc=North+Miami+Beach%2C+FL) | Page request blocked | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q28 | Google AI Mode | Miami | Neuromodulator injections | [miamiderm.as.me](https://miamiderm.as.me/?appointmentType=59781846) | uninspectable\_booking\_static\_shell | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q29 | ChatGPT | Miami | Dermal fillers | [skinassociatesfl.com/dermatology-doctors-miami](https://skinassociatesfl.com/dermatology-doctors-miami) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q30 | Gemini | Miami | Dermal fillers | [thecliniccompass.com/miami/best-med-spas](https://thecliniccompass.com/miami/best-med-spas) | Inspected static page | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Present | Present |
| Q31 | Google AI Overviews | Miami | Dermal fillers | [m.yelp.com/search](https://m.yelp.com/search?find_desc=Dermal+Fillers&find_loc=Brickell%2C+Miami%2C+FL) | Page request blocked | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |
| Q32 | Google AI Mode | Miami | Dermal fillers | [uclinicmiami.com](https://uclinicmiami.com/) | Inspected static page | Present | Unknown / not assessed | Present | Present | Unknown / not assessed |
| Q33 | ChatGPT | Miami | Laser skin resurfacing | [miamidermatology.com/cosmetic](https://miamidermatology.com/cosmetic) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q34 | Gemini | Miami | Laser skin resurfacing | [miamidermatology.com/cosmetic](https://miamidermatology.com/cosmetic) | Inspected static page | Present | Present | Present | Present | Unknown / not assessed |
| Q35 | Google AI Overviews | Miami | Laser skin resurfacing | [lumidermstudio.com/blog/microneedling-vs-laser-miami](https://lumidermstudio.com/blog/microneedling-vs-laser-miami) | Inspected static page | Present | Present | Unknown / not assessed | Present | Present |
| Q36 | Google AI Mode | Miami | Laser skin resurfacing | [4everyoung.zenoti.com/webstoreNew/services/aa5224a4-0aa9-47b6-a5a4-8b88fb7d6e63](https://4everyoung.zenoti.com/webstoreNew/services/aa5224a4-0aa9-47b6-a5a4-8b88fb7d6e63?serviceid=d19c23d1-9482-4b2f-8e48-7afdb7262dba&merchantId=aa5224a4-0aa9-47b6-a5a4-8b88fb7d6e63&booking_source=booknow&booking_medium=google) | uninspectable\_booking\_static\_shell | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed | Unknown / not assessed |

Of 36 predetermined slots, 32 selected a source (31 distinct URLs). The follow-up inspected 20 unique substantive static pages, represented in 21 slots. Among those 20 inspected pages: treatment detail 16, named-practitioner information 15, location information 12, contact information 18, and fee/payment context 10.

Slot incidence and unique-page counts are separate. A repeated selected URL was fetched and inspected once; procedure evidence is evaluated for the procedure in each slot.

True means a positive finding within the stated inspected scope. Null means unestablished, not absent; uninspected images, dynamic content, tabs, linked pages and entire websites are outside the inspection.

Named-practitioner information is not license verification or proof of a practitioner/branch/treatment relationship unless that relationship is explicitly documented.

Fee/payment context is not a verified treatment quotation. Publisher support contacts, navigation-only names and testimonial-only practitioner names do not establish the corresponding clinic evidence.

Answer narrative checks are bounded qualitative comparisons, not a provider-accuracy rate, clinical assessment or proof that a citation caused a recommendation.

Private raw answers and HTML/text snapshots are not redistributed. Public notes contain paraphrases rather than provider-name examples, full-answer text or body quotations.

The source-page follow-up describes this deterministic citation sample, not every page on a practice website. Missing page evidence and an inaccessible page remain distinct from a factual contradiction.

Practical interpretation

## Turn the evidence into an aesthetic clinic work list

For a clinic owner or marketing lead, the useful next decision is which service, practitioner and branch information to inspect—not whether one blended AI visibility number went up.

### Keep treatments and platforms separate

Use the [procedure-level table](https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study#aesthetic-intents) to organize neuromodulator injections, dermal fillers and laser resurfacing as different patient questions. Keep ChatGPT, Gemini, AI Overviews and AI Mode in separate reporting columns.

### Inspect the source behind the answer

Start with the [selected-source ledger](https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study#source-follow-up). On your own website, check whether the treatment page identifies the practitioner, actual clinic location and consultation route. Keep an inaccessible page separate from a page you have inspected.

### Recheck the same questions

The [repeat-source comparisons](https://rotgar.com/medical/resources/aesthetic-clinic-ai-visibility-four-platform-study#within-query-stability) show why a baseline needs more than one answer. Record the exact question, platform and date before using a later check to prioritize the next change.

### Choose the next practical step

Use the [aesthetic clinic AI visibility guide](https://rotgar.com/medical/niches/aesthetic-medicine/ai-visibility) for a page-by-page review. The [commercial service page](https://rotgar.com/medical/niches/aesthetic-medicine) explains audit, agreed implementation and repeat-check responsibilities.

Methodology

## How the aesthetic clinic panel was collected and analyzed

### Fix the aesthetic-clinic panel before collecting answers

The protocol and 18 exact query cells were fixed before paid collection. Three convenience-selected US markets were crossed with neuromodulator injections, dermal fillers and laser skin resurfacing, using ordinary-selection and practitioner/branch/evidence wordings. No practice brand was seeded in a query.

### Collect the four surfaces separately

DataForSEO LLM Scraper collected ChatGPT and Gemini. ChatGPT used force\_web\_search=true. The initial Google segment used Live endpoints. After repeated upstream technical failures, a prospective execution overlay admitted same-provider Standard queue delivery for remaining Google checks and four exact-input recoveries. AI Overviews retained depth 10 and asynchronous loading; AI Mode retained its separate Google surface. Responses retain actual delivery mode and collection time. US-English country-level location 2840, desktop and Windows settings were used; city names were explicit in the query.

Returned model labels — ChatGPT: gpt-6; Gemini: 3.5 Flash-Lite. Google AI Overviews and Google AI Mode did not return model-version labels. This collection does not establish equivalence to every signed-in consumer session.

### Keep repeats and collection states visible

The core contains 216 scheduled observation slots: three sequential same-day waves, 54 per platform. 5 diagnosed original temporary failures received separately admitted same-input recovery. Four Google failures were recovered through normal-priority Standard queue delivery before completion of the first wave. The separately admitted Live Gemini recovery occurred only after all original first-pass slots were terminal. Original and recovery modes and timestamps remain in the observation metrics. There were 221 core requests in total. 10 technical collection attempts are excluded: 6 preflight attempts across 4 fixed technical cells and 4 separately admitted diagnostic probes. 359 read-only queue GET calls retrieved already-posted tasks; they are not additional observation attempts.

205 final core slots contained an AI answer; 11 checks returned no AI answer; 0 retained an unresolved collection error. Original failure states and both timestamps remain in the public observation metrics. Scheduled repeat labels are not a claim that delayed recoveries occurred in their original chronological wave. Provider-side caching was not controllable.

### Count final-answer citations and source-set similarity

ChatGPT/Gemini use the provider’s declared final-answer sources, not unused search results. Google uses references inside top-level AI answer items, not ordinary organic listings, local packs or People Also Ask. URLs are deduplicated within an answer after hostname/www, tracking, fragment and trailing-slash normalization. Hostname means the normalized host, not corporate ownership.

Jaccard is intersection divided by union. Empty/empty citation sets are undefined, one-empty sets are zero, and both-nonempty summaries are reported separately. Within-query comparisons match exact query and platform; cross-platform comparisons match exact query and scheduled repeat label. Delayed recoveries are not represented as chronologically simultaneous captures.

### Inspect a fixed first-reference source sample

The qualitative sample uses the first plan-ordered selection wording, scheduled wave 1, then the first declared normalized citation. A same-input recovery of a temporary failure may fill that original slot later; the source ledger flags recovery and actual timestamps. No-answer, no-source and unresolved-error slots remain explicit. Dated source-page evidence supports only what was actually visible on that page.

The published codebook classifies URL navigation patterns conservatively. Treatment paths take precedence over generic location or practitioner prefixes. A path label is not a review of page quality or proof that a particular block caused a recommendation.

The public data omit full AI answers, API identifiers, credentials and private ClinicTruth project information. They retain prompt cells, observation metrics, citations, pair inputs and review evidence needed to inspect the analysis.

### Interpretation and limitations

- This is an exploratory same-day panel of 18 convenience-selected query cells, not a representative estimate of all US aesthetic-clinic searches or practices.
- The two wordings request different information; their differences are not an isolated causal prompt experiment.
- Cities were named in the query. Physical city IP, signed-in personalization and provider-side caching were not controlled. Delayed missing-only recoveries widen the window for their original repeat slots.
- Citation overlap describes visible reference sets, not provider quality, clinical suitability, patient demand or the hidden causal source of a recommendation.
- URL-path categories are conservative navigational proxies. Selected source-page facts do not establish that a website feature causes AI inclusion.
- No website intervention was applied. The practical aesthetic-clinic guide combines observed findings with Rotgar’s interpretation rather than claiming proven ranking factors.

Download and reuse

## Download the public v1.0 research materials

Inspect the prompts, denominators and source sets behind the findings. The public package is available without an email gate under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Full AI answers, API credentials and internal collection identifiers are not redistributed.

[text/csv **Platform results · CSV** Answers, citations and repeatability, with exact platform denominators.1.1 KB `sha256:0ec0beb0878da429e1bedddc83b10b138921c7bd8e7f30c2e7ce6503bc68d6e9`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/platform-summary.csv) [application/json **Complete public data · JSON** The sanitized observation metrics, citations, pair tables and source follow-up in one file.1.1 MB `sha256:394b2cd2b9eff0df23965430608804ed326287900be88e3a1336ce86bc76edb3`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/aesthetic-ai-visibility-four-platform-v1.0.json) [text/csv **Exact aesthetic-clinic questions** All 18 fixed city, nonsurgical procedure and wording cells.2.9 KB `sha256:773797222d8751c60ff4d1faf69fcc9ae3c3f73f35e0b1a02942931c45d22946`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/prompt-manifest.csv) [text/csv **Observation metrics** All 216 planned core observations, including no-answer and collection-error states.37.1 KB `sha256:64dcf94bf9d42e84ae8d72a79f95788f616aa0ffadc95e93c8bfb05e88fca3f3`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/response-metrics.csv) [text/csv **Deduplicated answer citations** Normalized answer references and conservative URL-path categories.241.9 KB `sha256:be9cd67add83bd24a43d9b1bb59a7510e0392b4897747cf82489447675de86a6`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/citations.csv) [text/csv **Repeat-source comparisons** All 216 within-query pairs and the source-set eligibility rules.25.8 KB `sha256:f262e7794e18cd6104bd124c0cdf5a7ec8b9aa101f931661a4b37490ec3caeed`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/within-pairs.csv) [text/csv **Cross-platform source comparisons** All 324 matched pairs across the six platform combinations.38.7 KB `sha256:010c10f8ab4f550af5a04e52fc25e35acaf186df83d0430a5ce19d2205452bb8`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/cross-platform-pairs.csv) [text/csv **Cited page types** Platform-level navigational URL categories; not a ranking-factor score.2.2 KB `sha256:f1ca951f8bd85616562a498e18abd72373ed342979f1bd3090a9494b76c0e0f5`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/source-types.csv) [text/csv **Source-page follow-up** All 36 predetermined first-reference slots and their review evidence.45.2 KB `sha256:2a47e4119c91aed6fb43dbee6e9c12fb4188c3d027ff906bcb442051b81bd015`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/qualitative-source-review.csv) [text/markdown **Methodology and definitions** Collection, exclusions, denominator choices and interpretation boundaries.6.4 KB `sha256:62a001152d726a7d61108c48b3837429101b5d3468ddc432e15f0ab61d2d70d2`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/METHODOLOGY.md) [text/plain **Reproduce the aggregate checks** Run the dependency-free Node.js checks against the sanitized public tables.9.1 KB `sha256:779b615a50a96f498c05ad064556e79acb9b3a1ee43a6c51ced4ecc3333d719c`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/reproduce.mjs) [text/plain **Citation metadata** Original v1.0 metadata; use the version DOI in the research record for current citation.442 B `sha256:61a132f73c46e00ae71cbc32944b96ab55fb5c4eefd1941e7e51eb262ba61c8e`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/CITATION.cff) [application/json **File manifest** Public payload inventory, byte lengths and SHA-256 checksums.5.6 KB `sha256:553cc609d0f0574c8fd08b4414274c012cb233ff9c4a161798628ce7d852f8d3`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/manifest.json) [text/plain **Verify the package** Checksums for every payload and the manifest.2.2 KB `sha256:c9cab5c2df4b5876ec08bce6f303023eaed274a10ce31e5a40ac92a555ddca26`](https://rotgar.com/research-data/aesthetic-ai-visibility-four-platform-v1.0/SHA256SUMS.txt)

Citation and research record

## A separate, versioned aesthetic clinic dataset

> Yudin, E. (2026). Aesthetic Clinic AI Search Visibility Across Four Platforms (Version 1.0) \[Data set\]. Rotgar Research. https://doi.org/10.5281/zenodo.23693457

Author

[Evgeniy Yudin](https://rotgar.com/about), Founder and Strategy Lead · [ORCID](https://orcid.org/0009-0007-8400-9561)

Publisher

Rotgar Research

Version DOI

[10.5281/zenodo.23693457](https://doi.org/10.5281/zenodo.23693457)

Zenodo record

[Open the archived dataset ](https://zenodo.org/records/23693457)

Public dataset mirrors

[GitHub release v1.0](https://github.com/evgeniy-yudin/aesthetic-ai-visibility-four-platform/releases/tag/v1.0) · [Kaggle dataset](https://www.kaggle.com/datasets/evgeniyyudin/aesthetic-clinic-ai-search-visibility-4-platforms) · [Hugging Face dataset](https://huggingface.co/datasets/RotgarSett/aesthetic-ai-visibility-four-platform)

Published

October 27, 2026

Fieldwork window

2026-10-27T12:42:24.960Z – 2026-10-27T13:25:28.963Z

Version and license

1.0 · [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)

### Trace the evidence

One Gemini Live timeout received a separately admitted same-input recovery after all original first-pass slots were terminal. The admission SHA-256 is ee9d006ab4f8d165cc2b7313480e76b1715918ece157096aac4da8e970ab4ab8. This is an additional collection attempt, not an independent observation.

A prospective execution overlay changed Google delivery from Live to same-provider Standard queue while preserving every scientific query and market setting. Its SHA-256 is cc6e759ebe788bceeb0065e61a9cb92ff393e484cfc092997558b69ceaa9dee5. Response metrics retain delivery mode, scheduled repeat label and actual timestamps.

The research artifact hash identifies the sanitized public analytical JSON containing the core observation metrics, citation table, repeat pairs, matched platform pairs and source-review slots. The fixed protocol hash is 33c51d182c85fe3b5a11c244e534b3c0eb93ba0d7d4d0c700e1d23a0937ad330.

This is a new nonsurgical aesthetic-clinic dataset. The earlier Aesthetic Medicine & Medspa source-overlap subset is separate historical evidence. IVF, dentistry, bariatric or cosmetic-surgery results are not relabeled as medspa findings.

Version 1.0 is archived in Zenodo under the version-specific DOI shown in this research record. The original public data package remains unchanged; use that DOI for current citation.

**Research artifact SHA-256** `394b2cd2b9eff0df23965430608804ed326287900be88e3a1336ce86bc76edb3`

This aesthetic clinic panel has its own dataset identity. Earlier Rotgar studies retain their original populations, dates and publication records.

Continue from the evidence

## Aesthetic clinic actions and related research

AI Visibility Audit

## See how AI answers describe your aesthetic clinic

The AI visibility audit starts with one clinic or selected location, one priority market and one patient language.

**Best fit** Clinics, hospitals and medical groups in:

- Dental
- Aesthetic medicine
- Physiotherapy
- Fertility & IVF
- Dermatology
- Hair transplant
- Cosmetic plastic surgery
- Mental health
- Multi-specialty groups

- Your current visibility and named competitors
- The sources AI answers cite
- A prioritized list of improvements to implement

Need to cover several clinics, locations, countries, markets or patient languages? [Book a 30-minute call ](https://calendly.com/z-rotgar/30min) We’ll discuss the scope for your organization.
