Starter pack · Clinics & healthcare
A clinic where no booking is lost, no patient is forgotten, and the doctor writes nothing down
Four AI agents for small and mid-sized aesthetic, dental, physiotherapy and specialist clinics. They take appointments, send reminders, follow up after treatment, answer questions about your services, and draft notes for the doctor to check — while everything to do with treatment stays in the hands of the doctors and nurses. Pre-assembled, live in 2–3 weeks.
Work in
Clinics & healthcare+ human approval
To the customer
A person approves before it ever reaches the customer· drag to rotate
A fit if you…
- Clinics with 1–3 branches and 1–5 doctors, where admins answer LINE and take appointment calls by hand all day
- Patients missing appointments, reschedules clashing, and nobody following up after treatment — so they do not come back
- You want to use AI but worry about patient data and PDPA, so you have not started, and you do not want to pay for an Audit first
A day with this pack running
- 07:30The agent gives the head nurse today's list: who comes at what time, for what, who has not confirmed, and which post-treatment cases reported symptoms overnight
- 09:15A patient messages on LINE for a filler appointment on Saturday. The agent checks the doctors' schedule, offers 2 slots, confirms the booking, and sends the preparation notes
- 11:00A patient asks about laser course prices and this month's promotion. The agent answers from the price list the clinic approved. Asked “what should someone with my skin have done?”, it passes the question to a nurse
- 13:00Tomorrow's reminders go out. Anyone asking to move is rebooked on the spot, and the slot that opens is offered to the people on the waiting list
- 17:30The doctor finishes clinic. The agent drafts the case summary from the treatment notes for the doctor to check and confirm before it enters the clinic's records
- 20:00It asks patients treated 3 days ago how they are. Normal answers close the case; an unusual answer, such as more swelling or redness, reaches the nurse on duty at once
The output you get
Patient
Could I move my Thursday appointment to Saturday?AI
Yes — Saturday the 14th has 10:30 and 15:00 free. Which time suits you?Health data kept under PDPA · anything needing a diagnosis always goes to a doctor
An illustrative example — the names and figures are invented so you can see the shape of the work; the real thing is built around your business.
Example: Clinics & healthcare
LIVE DEMO1 · Work arrives
2 · The AI agent does it
- Booking, reminders, confirmations
- Post-treatment follow-up
- Service, price and promo questions
- Notes drafted for the doctor
3 · You just approve
Procedures that need an assessment first, and slots that clash with time a doctor has blocked, always go to the head nurse to confirm.
Result
-38%
patients missing appointments
What is in the pack
4 AI Agent already assembled
Takes bookings from LINE and by phone, checks the real doctor and treatment room schedule, confirms the slot, sends reminders 1 day and 2 hours before, handles reschedules and cancellations, and fills the gap from the waiting list automatically.
Procedures that need an assessment first, and slots that clash with time a doctor has blocked, always go to the head nurse to confirm.
Checks in with patients on the rounds the doctor sets, such as 1 day, 3 days and 7 days after treatment, using the question set the nurses approved. Normal answers close the case and book the next visit; an unusual answer reaches a nurse with the history attached, immediately.
The AI does not suggest medication, aftercare or treatment, in any case. Every unusual symptom has to be answered by a nurse or a doctor.
Answers questions on services, prices, promotions, opening hours, how to prepare and how to get there, from a knowledge base the clinic edits itself, 24 hours a day on every channel, and can open the appointment in the same conversation.
Anything that amounts to a diagnosis, a comparison of treatment results, or a discount outside the terms, is passed to a person every time.
Drafts the case summary from the treatment notes or the doctor's dictation, in the record format the clinic uses, and brings up drug allergies and the previous treatment before the consultation.
Every summary is only a draft. The treating doctor has to read it and confirm it before it enters the records — the system files nothing automatically.
All included
What is in the box
- An appointment schedule linked to the doctors' calendars and the real treatment rooms, with a waiting list and the clinic's rules for moving appointments
- Reminder messages, preparation notes and post-treatment question sets, split by procedure and editable by you
- A knowledge base of services, prices and promotions that admins edit themselves without waiting for us
- An approval screen for nurses: unusual symptom cases, appointments that need an assessment, and drafts waiting for a doctor to confirm
- A PDPA data map, a consent form over LINE, a retention and deletion policy, and a log of access to patient data
- A weekly dashboard (bookings, attendance, post-treatment cases, questions the AI could not answer) and 2 training sessions for admins and nurses
Go live
Live in 2–3 weeks
Step 1 / 3 · Collect the clinic's rules and set up PDPA
Week 1
- 1Week 1
Collect the clinic's rules and set up PDPA
Two to three hours with the owner, the doctors and the admins to collect the doctors' schedule, the service and price list, the follow-up rounds, and the line the AI must not cross, plus the data map and the consent form.
- 2Week 2
Connect the systems and test on real chats
- 3Week 3
Go live branch by branch and tune
-38%
patients missing appointments
from the clinics we have done, within 2 months of go-live, because of two reminders and rescheduling in the chat
3 hrs/day
given back to admins
from no longer handling bookings and repeat questions by hand, spent on the patients in front of them instead
+22%
patients coming back for more treatment
from following up after treatment and booking the next visit on the spot; usually visible in month 2–3
Price and scope
One price, paid once, to install the whole pack, plus a small monthly fee for models and keeping the system running. All in, about the same as one part-time admin per month. Fixed scope as set out in the pack, with no hidden costs.
Not included
- Building a new records system (EMR/HIS), or migrating everything out of the old one
- A full Voice Agent taking phone calls (can be added in month 2–3)
- Marketing, content or advertising, and AI analysis of medical images
Where you go when you grow
Once you have more than 3 branches, or want to connect to a complex records system, move up to Agentic Workflow Design and Implement & Coach built around your clinic, or add the Customer service on every channel and Operations and back office bundles, without starting over.
Safety in this business
- The AI does not diagnose, does not suggest medication, and does not recommend treatment, in any case. Every question about symptoms goes to a nurse or a doctor, and unusual cases reach a person immediately rather than waiting for the next round.
- Patient data is sensitive data under PDPA. We build a data map, take consent before collecting, set how long data is kept and how it is deleted, and log who accessed what and when.
- Treatment history is not sent out to train models. The model sees only the data that task needs, and you can choose a model that runs in Thailand or a Private LLM when the clinic wants it.
FAQ for Clinics & healthcare
Will the AI answer questions about symptoms, or recommend treatment to patients?
No. The AI answers only on services, prices, appointments and the preparation steps the clinic has approved. Questions about symptoms, diagnosis or comparing treatment results are passed to a nurse or a doctor every time, and we draw that line together with your doctors in the first week.
How safe is patient data, and does this breach PDPA?
Health data is sensitive data that needs clear consent. This pack comes with a data map, a consent form, a retention and deletion policy, and access logging. The data is not used to train outside models, and if the clinic wants it, you can use a model that runs in Thailand or a Private LLM.
We already use a queue or records system. Do we have to change it?
No. We connect to the calendar and the systems you have, where they offer an API or can export data. If the old system cannot connect at all, we run the pack's appointment schedule alongside it, and admins still see everything in one place.
What if a patient wants a real person, or the AI gets it wrong?
A patient can ask for a person at any time and the system hands over to an admin with the conversation summarised. Every AI answer is logged, and there is a weekly report of the questions it could not answer, so your team can fix the knowledge base itself straight away.
How do we start, and what do we need to prepare?
Start with a free 30-minute call, no obligation, to see whether this pack fits your clinic. Then have the doctors' schedule, the service and price list, and 1–2 months of past chats ready. We handle the rest within 2–3 weeks.
Cases from similar businesses
Packs for other businesses
A free 30-minute call to see if the clinic pack fits you
Tell us how many doctors you have, what queue system you use, and where the bookings are going. We will say straight out whether this pack is enough or whether it should be designed around you. No obligation.
Want to see it working first? Open the sample workspace for a business like yours — this exact team of agents is running in it. Sample data, no sign-up.