Innoderm is a private aesthetic medicine clinic in West London with two consulting rooms, one doctor, and a brand built over four years on referrals. The waiting list was twelve weeks. The bookings system was the doctor herself, answering messages on her phone between consultations.
By the end of 2023, the maths stopped working. The clinic was billing under capacity, the doctor was working sixty-hour weeks, and every attempt to introduce a "real" booking system had failed within three weeks. Patients did not want a portal. They did not want to download an app. They wanted to keep messaging the number they already had.
What we did
The first two weeks were spent watching, not building. We sat with the doctor through a normal week of patient interactions. We mapped where time was lost. The pattern was clear: the bookings system was not broken. The fulfilment system was.
Patients messaged in WhatsApp. The doctor wrote slots in her diary. The diary was not visible to her receptionist. The receptionist double-booked twice a week. Each double-booking cost a forty-five minute apology call, a fifty-pound apology cream, and a slot of the doctor's next-available time.
We did not replace WhatsApp. We kept WhatsApp as the surface. Behind it, we built a Postgres source of truth, a Next.js admin for the receptionist, and a WhatsApp Business API integration that confirmed every booking in the patient's existing thread within eight seconds. The doctor's phone became a read-only view of the same calendar.
What changed
Online bookings tripled in the first nine months. More importantly: zero double-bookings since week three of go-live. The receptionist now spends an hour a day on bookings, down from four. The doctor's working week is back to thirty-eight hours.
The revenue effect was secondary to the load-shed. The clinic was already at near-capacity for in-person consultations. What the system unlocked was a structured pipeline for follow-ups, package renewals, and treatment-cycle reminders — work that was previously falling through the cracks because nobody had time to surface it.
By month nine, the clinic had hired a second doctor and was running both rooms full-time. The bookings system handled the second doctor on day one without any changes — the architecture was multi-practitioner from week one, even though we only needed one.
What's next
A third room is opening in Mayfair in Q3 2026. The booking system will extend with no additional engineering work. We continue to run maintenance, security patches, and quarterly product reviews on a 24/7 on-call rotation. P1 response is fifteen minutes, including weekends.
The clinic stopped asking us for new features eleven months in. They ask us for fewer things now — that is the metric we are most proud of.
“Bookings doubled the day we moved the confirmation step into the WhatsApp thread the patient already trusted.”