The situation
A Singapore medical group running four clinics had a catalogue most businesses would recognise: 1,062 services and prices — health screenings, vaccinations, diagnostic tests — spread across systems and changing constantly. Patients couldn't easily find what they needed or what it cost. Staff answered the same questions all day. The knowledge existed; it just wasn't usable.
What we built
We made the group's own catalogue answerable in plain language. A patient can ask the way they'd ask a person — "how much is a full health screening with the heart add-on?" — and get an accurate, current price drawn from the group's live data. We connected it to their clinic management and booking system, so it doesn't just answer — it moves the patient toward booking. It runs where patients already are: WhatsApp and the website, around the clock.
The assistant patients talk to is the visible part. The work underneath is the point: taking messy, high-volume, frequently-changing knowledge and making it reliable enough to put in front of customers — and to act on.
Built to be trusted
In healthcare-adjacent work, "mostly right" isn't good enough. The system is built with accountability and audit baked in, aligned to frameworks like AIHGle 2.0 — so answers are traceable and the group stays in control of what it says.
In production today
- 1,062 services and prices — answerable instantly, in plain language
- 4 clinic locations served
- 24/7 enquiry coverage
- 3 systems connected
- 30–50 patient conversations a day
The capability this proves. This wasn't a chatbot project — it was a knowledge problem. The same capability applies to any business sitting on a large, messy, fast-changing body of its own information that customers need and staff keep re-explaining. The medical group is one example; the capability travels.