Oris

Med spas / AI visibility & content

Ask ChatGPT about your clinic. You may not like the answer.

Patients increasingly ask an assistant before they ask Google. If your hours, service area or treatment list are not published as facts a machine can read, it will improvise — or name the clinic down the road instead of you.

What it costs
from $1,000/mo
What arrives monthly
4–8 pieces, one answer report
Who it is for
US clinics, one location up
Commitment
30 days notice
Same clinic, same question

“Does this clinic treat patients outside the city?”

Before

They appear to serve clients across the United States. Hours are not listed — you may want to call to confirm.

After

They treat patients in the Dallas–Fort Worth area only. Open Tue–Sat, 9am–6pm, with consultations bookable online.

A real pattern we have fixed on a live store: the assistant claimed nationwide service when the business served one state. The fix was publishing the fact, not arguing with the model.

$5K

lifetime value of one retained med spa patient, at the top of the range

Industry range commonly reported at $2,000–5,000. We use the low end in every calculation.

A wrong answer is not a missed click

It is a patient who booked somewhere else, believing something about you that is not true. And unlike a bad ad, it keeps happening quietly — the assistant will repeat it to everyone who asks, every day, until the underlying facts change.

These are published industry benchmarks, not Oris results. Your own numbers come out of the free audit.

What we run

Six pieces, and one of them is just checking

Most of this work is unglamorous: publishing facts in a machine-readable form and killing the contradictions that make assistants hedge.

The fact layer

Hours, service area, treatments, price ranges, providers and policies, published as structured data instead of buried in a PDF or an image.

Answer monitoring

Every month we ask the assistants the questions your patients ask and log what comes back — including when it is wrong.

Content that answers

Four to eight pieces a month built around real questions, written so a machine can quote a paragraph without mangling it.

Schema and technical

Organization, LocalBusiness, Service and FAQ markup, clean crawl paths, and no accidental blocking of the assistants you want reading you.

Contradiction cleanup

Old directory entries and stale pages that state different hours or a different address are why assistants hedge. We hunt them down.

Reporting on answers, not rankings

What each assistant says about you this month versus last, and which corrections landed.

What it costs

One number, and what moves it

$1,000 a month, month to month. No long lock-in — 30 days notice, any time. The first corrections are usually live in about two weeks, because most of them are facts you already know and have never published.

The whole price list

from $1,000/mo

What moves it
How many pieces a month and whether we also rebuild your existing pages for AI answers.
Not included
Paid distribution. We do not buy links.
Commitment
$1,000 a month, cancellable on 30 days notice. No setup fee.

None of this work touches patient records. Everything we publish is what you already tell any patient who calls.

Where we are honest with you

Three things we will not promise you

Every vendor in this category is selling one of the three. We would rather lose the deal here than in month four.

  1. We cannot get you ranked inside ChatGPT

    There is no ranking to buy, no submission form, no position to hold. Answers shift by account, location and model version. What we control is what an assistant finds when it looks for you.

  2. Nobody can measure citations reliably

    No major assistant publishes citation or impression data, and Search Console folds AI Overview impressions into your ordinary totals. Every share-of-voice dashboard on the market samples prompts and extrapolates. We run the test by hand and show you the answers instead.

  3. This is not your main channel yet

    For a single-location clinic the realistic prize is a thin long tail of referrals — plus not being misrepresented to the patients who do ask. Anyone selling you this as a growth channel has skipped that sentence.

Why do it anyway

A wrong answer costs you patients today. Fixing it is a one-off content job, not a permanent spend.

Why it compounds

The same structured facts strengthen your local ranking and your site’s ordinary SEO at no extra cost.

Why now

Almost no clinic in your market has done it, so the cost of being first is unusually low.

You do not need us for this part

Run the test tonight, for free

Ten questions, four assistants, twenty minutes, signed out of everything so your own history does not skew the answers. Log what comes back verbatim and flag what is wrong. That is the entire method, and the guide hands you all ten.

If you have a spare weekend, do it yourself — that is not a rhetorical offer. What you pay us for is doing it every month and fixing what it finds.

The full guide, with all ten questions
  1. Best med spa in [your city]
  2. How much does [your top treatment] cost in [your city]
  3. What are [your clinic name]’s hours
  4. [Your clinic name] vs [your closest competitor]
  5. …and six more, in the guide

Questions about AI visibility

Straight answers

Is this just SEO with a new name?

No, and treating it that way is why most agencies get it wrong. Classic SEO optimises a page to rank; this optimises facts so an answer engine can state them correctly. A clinic can rank fine in Google and still have ChatGPT give patients the wrong hours, the wrong service area or the wrong price range. Local SEO fights for the map pack. This makes sure that when an assistant summarises your clinic, it says the right things and names you rather than a directory.

How would we even know what AI says about us?

We ask it — systematically, across the questions your patients actually ask, on the assistants they actually use, and we log the answers. That baseline is part of the free audit, and it is usually the moment the conversation gets serious.

What if the answers are just wrong?

That is the common case, and it is fixable. Wrong answers come from thin, contradictory or missing structured facts. We publish the correct ones in a form machines read, then re-test until the answer changes.

Can you guarantee ChatGPT will recommend us?

No, and anyone who does is selling something they do not control. Answers move with the account, the location, the model version, and sometimes between two identical prompts an hour apart. What can be committed to is the input side: correct, complete, consistent, machine-readable facts. That part is engineering and you can verify it in an afternoon. The ranking is not ours to sell.

Does any of this bring patients, or is it a vanity metric?

Referral traffic from assistants is still a thin tail for a local clinic, so we do not sell it as your main channel. It matters because the wrong answer actively costs you patients today, because it is cheap to fix once, and because the same structured facts strengthen ordinary local ranking at the same time.

We are one location with a small team. Is this worth it at our size?

Usually more than it is for a group, and most of it is a one-off cleanup rather than a permanent spend. One location means one set of hours, one address and one provider list to make consistent — the same job as a ten-location group without the multiplication. If your phone is not being answered, fix that first; a patient an assistant sends to a clinic that rings out is a patient you paid attention to lose twice.

Does this touch patient data?

At no point. Everything we publish is information you already give any patient who calls: hours, service area, treatments, price ranges, who performs them, and your policies. Nothing here needs access to your records or your practice management system. Photos, reviews and testimonials are a separate decision with their own HIPAA, FTC and state rules — we are engineers who work to documented practice, not lawyers, and we will say so before you ask.

Why trust a team outside the US with this?

Oris is the clinic practice of HeadPills, running since 2021 out of Wroclaw, Poland, with more than a hundred businesses shipped across Europe. This particular work is location-independent — it is your facts, your markup and a test anyone can re-run. We have no US clinic clients yet, which is why there is no case study on this page. Run the ten-question test yourself and you can check our method without hiring us.

What does it cost?

From $1,000 a month: structured fact layer, four to eight content pieces, schema work, and monthly answer monitoring across the main assistants. How many pieces a month and whether we also rebuild your existing pages for AI answers. No long lock-in — 30 days notice, any time.

Free clinic audit

We ask the assistants about your clinic first

The free audit includes what ChatGPT, Perplexity and Google's AI answers currently say about you — hours, service area, treatments — and where they are simply wrong.

  • Findings in two business days
  • No call required to receive them
  • Yours to keep either way

No spam, no list. One reply, from a human, within two business days.

Get the free audit