Skip to content
For private clinics

Reception cannot answer WhatsApp during a consultation.

Appointment requests arrive on WhatsApp, Instagram and the site form while your staff are with patients. They get answered when someone is free, which is usually after the caller has booked somewhere else.

Book a call
See it runNew patient enquiry on WhatsApp
WhatsApp message
Trigger
New or returning
Checked
Slots offered
Your calendar
Booked
Confirmed
Reminder scheduled
Day before

Read, triaged and booked in 4.8 seconds, with the reasoning stored.

Where the appointments actually go

A clinic loses appointments in the gap between the request and the reply, and not to a better clinic. To whichever one answered first. The messages arrive on three channels, none of them is anyone's job, and the person who could answer is in a room with a patient. By the evening there are eleven unread threads and no way to tell which of them were new patients.

Four things we automate for clinics

Each one is a workflow you own, running in your own accounts, monitored and documented.

Appointment requests, answered in two minutesA request from WhatsApp, Instagram or the site form gets a reply offering slots your calendar really has. The patient picks one and it is booked. A person is one tap away at any point in the conversation.
Reminders that cut no-showsA confirmation when the appointment is made, a reminder the day before, and a rescheduling link in both. No-shows are the cheapest thing a clinic can fix, and almost nobody fixes them.
Recall for treatments that repeatHygiene visits, follow-ups, annual checks. The system knows who is due and asks them, on the channel they used last time, without anyone keeping a list.
Triage before the desk sees itNew patient or returning, which treatment, which practitioner, urgent or routine. The enquiry reaches your desk already sorted, so the first human minute goes on deciding rather than reading.
What ships
  • WhatsApp Cloud API on your own clinic number, never a shared one
  • Slot offers read from your real calendar, never invented
  • One-tap handoff to a person, with the whole thread visible
  • Reminder and rescheduling flow with opt-out handling
  • A runbook: what breaks, how to recover, what it costs to run

Questions clinics ask

Does the assistant see patient records?
No. It handles appointment logistics only: who is asking, which treatment, when they can come. We do not connect to your clinical system and we do not process health records. If a conversation moves toward symptoms or clinical advice it hands over to a person, and says so.
What stops it inventing an answer?
It answers from a list of approved responses you write, not from a model guessing. Anything outside that list becomes a handoff to your desk instead of an improvised reply. You read every answer it is allowed to give before it goes live.
What happens if a patient writes something urgent?
Urgency routing is part of the setup. You define what counts, and those messages skip the queue and alert a person immediately rather than being handled by the assistant.
Is this GDPR compliant?
The data stays in your own accounts, we sign a data processing agreement, and Gate 1 of every project includes a written GDPR review of what is collected and how long it is kept. We do not touch clinical records at all, which keeps the scope narrow on purpose.
How long before it is running?
Five to eight working days from discovery to a signed spec. The build depends on how many channels you connect. You get a fixed scope, a fixed price and a fixed date before anyone builds anything.

Tell us the process you hate most.

Thirty minutes, no deck. We walk one process end to end and tell you whether automating it pays for itself.

Book a callNo newsletter. No sequence.