Elase Med Spa

What to do next

Ranked by urgency. Every recommendation below traces to a figure elsewhere in this report — nothing is added to fill a slot, and where a finding is not supported, no card appears.

---

Reputation movement

Captured 2026-08-19 to 2026-08-20

Where you stand today, and how fast that is changing. Rating and volume are a single capture; the rate below is measured over 90 days from review dates.

Across the 55 nearby clinics where review text was collected, 27 received a review in the last 90 days, at a median of 2. The fastest of them, Aesthetic and Clinical Dermatology - Hinsdale, added 20. You already sit in the local top quartile on review volume; the position to defend is the rate, not the total.

You have replied to 2 of the 34 reviews we hold for you (6%). The median across the 43 clinics here where we hold enough reviews to compute a rate is 54% — 5 answer every one and 5 answer none, so unlike the rating this is a number that still separates clinics here. A reply is visible to the next person reading, and it is the only thing in this section you can change this week without spending anything. We do not claim it lifts your rating: across the clinics we measured, the relationship between replying and star rating is weak.

Google review count, you against your nearest 12. Your clinic is marked in green. Month-over-month movement is not in this edition — every clinic here has been captured once, over a three-day window in August. The second capture is what turns this section from a position into a trend.

When your market is open

Listing completeness

What your market publishes

---

How you show up in AI answers

Captured 2026-09-22 to 2026-09-28

Patients increasingly ask an assistant “best med spa near Hinsdale or Oak Brook” instead of searching. The answer names three or four businesses out of the 234 here. This is which ones, and how often.

You are named in 20 of 60 answers, and by all 3 sources tested — 3 AI assistants. The engines disagree sharply about you. Gemini names you in 55% of its answers; ChatGPT in 10%. That gap is the finding — these sources read different indexes, so being visible in one says nothing about the other. Those answers name Elase Med Spa as a brand. 3 of your locations are reached by the same searches, and the answers do not say which one they mean, so the count is for the name, not this door.

Claude 7 of 20 · Gemini 11 of 20 · ChatGPT 2 of 20 — each source on its own question set, repetitions and date, listed below. They are not one capture and are not averaged into a single figure.

Google AI Mode is a different kind of source — Google’s conversational search, which a patient opens deliberately. It is counted on its own and never added to the assistant answers above: in Hinsdale we ran 30 searches on 2026-09-28; 30 came back; Google showed an AI answer for 30 of them, and 20 named you; in Oak Brook we ran 30 searches on 2026-09-28; 30 came back; Google showed an AI answer for 30 of them, and 2 named you.

Google AI Overview is a different kind of source — the block above Google’s own search results, which Google shows for some searches and not others. It is counted on its own and never added to the assistant answers above: in Hinsdale we ran 30 searches on 2026-09-28; 28 came back and 2 failed at the data provider, which says nothing about you; Google showed an AI answer for 18 of them, and 10 named you; in Oak Brook we ran 30 searches on 2026-09-28; 29 came back and 1 failed at the data provider, which says nothing about you; Google showed an AI answer for 13 of them, and 2 named you.

Google’s AI Overview in this edition: it is not measured for Hinsdale and Oak Brook in this capture — the Overview there was captured on 2026-09-28 and the assistants above were asked on 2026-09-22, which is two captures, not one, so the two are not compared. This section reports the sources it asked and makes no claim about Google’s own AI answers for those towns.

Question by question

The 10 questions the assistants were asked about each town in your competitive scope. An average across them hides the only thing worth acting on: which question you are absent from. Worst first.

Named in is how many of the answers to that exact question mentioned you at all; best position is the highest place you reached in any one of them. Both are given because they are different facts — being named once at the top is not the same as being named four times near the bottom. Where no clinic took first place more than once, the assistants disagreed with each other and no leader is named.

Who they name instead

What this is, and what it is not. These answers come from each vendor’s API, not from the consumer app. The app adds its own system prompt, your own history, and personalisation we cannot see, so treat this as directional evidence about how a clinic surfaces in a grounded assistant answer — not as a measurement of what any one patient sees. Assistant answers also vary run to run, and on current models the settings that would pin them down are removed outright, so every prompt was asked more than once wherever the capture allowed it, and the metric is how OFTEN you were named, never whether you were. Each source as it was actually measured: Claude - prompt set v1, 10 question(s) per town asked 1x on 2026-09-22; Gemini - prompt set v1, 10 question(s) per town asked 1x on 2026-09-22; Google AI Mode - prompt set v1g, 30 question(s) per town asked 1x on 2026-09-28; Google AI Overview - prompt set v1g, 30 question(s) per town asked 1x on 2026-09-28; ChatGPT - prompt set v1, 10 question(s) per town asked 1x on 2026-09-22. The model is part of the instrument, so each is pinned and named: Claude claude-opus-5, Gemini gemini-3.7-flash, ChatGPT gpt-5.2.

Only clinics already in this market map can be recognised. An assistant naming a business we have never collected is not counted, which understates competition rather than inventing it. 41 of the clinics we hold cannot be recognised in prose at all — their names are pure initials or pure industry words — and “not named” and “not detectable” are different facts.

---