
Dental Clinic Booking Automation with n8n
A clinic loses a patient the moment an appointment request goes nowhere. The phone rings after hours, a Facebook message sits unread, or a form sends an email nobody checks. For Shifa Advanced Dental Care & Implant Center in Tongi, we built a booking path designed to put each request in front of the doctor within moments, and to tell us if anything goes wrong.
This post explains how that booking flow works. For the full project (the bilingual website, search setup and Google Business Profile work), see the Shifa Advanced Dental Care case study.
What the clinic needed
- A way for patients to ask for an appointment outside phone hours
- A record of who asked, kept by the clinic, not by us
- An alert the doctor sees on his phone, with the patient's number ready to call
- No patient data stored on the website itself
How the booking flow works
The website form sends each request to a webhook in n8n. From there, the workflow runs in this order:
- Check the key. The form sends a shared secret key with each request. Requests without the right key are refused, so fake or scripted submissions do not reach the clinic.
- Tidy the details. The booking details from the form are put into one consistent format.
- Check the request is complete. An incomplete request gets a clear error back instead of being saved.
- Save to the clinic's sheet. The booking is added as a new row in a Google Sheet in the clinic's own Google account.
- Confirm to the website. The patient sees that the request was received.
- Alert the doctor. A Telegram message goes to the doctor with the booking details. He can tap the number to call the patient back. A copy goes to ByteFlow AI Labs so we can see the system is working.
What happens when something fails
Most automations are built for the happy path. The part that matters to a clinic is what happens when the save fails, for example if Google is briefly unavailable.
In this workflow, a failed save takes a separate path: the website gets a failure response, and ByteFlow AI Labs gets an urgent Telegram alert with the request details, so the patient can still be called back. A booking does not disappear quietly.
Privacy by design
- The website stores no patient data. Bookings go straight to the clinic's sheet.
- Bookings are deleted weekly and kept no longer than 14 days.
- Analytics loads only after the visitor accepts cookies (Google Consent Mode v2), and its data is kept for 2 months.
- The clinic owns the Google account, the sheet and the Telegram bot.
How we tested it
Before launch, we ran the full path (form, sheet and both Telegram alerts), the failure alert, and the rejection of requests without the key. The first real patient booking arrived on 14 November 2026: it was saved and both alerts were sent.
We will add booking and search figures after the clinic's first monthly report.
Could this work for your clinic?
The same pattern fits most clinics that take appointment requests by form, WhatsApp or phone: one place to record requests, one alert the right person sees, and a plan for when something breaks. The tools can change (a CRM instead of a sheet, WhatsApp instead of Telegram), but the order of checks stays the same.
If enquiries at your clinic get lost between the phone, Facebook and the front desk, tell us how your bookings work today and we will tell you honestly what is worth automating.