AI customer assistant for healthcare providers: appointment questions answered, under the stricter policy
How a clinic configures Kav: appointment and billing lookups from its own scheduling system, with health data held to the strictest policy, plus a worksheet.
Most of what a clinic's front desk hears is logistics, not medicine: when is my appointment, can I move it, where do I go, what do I owe. It arrives in the same hours the desk is busiest.
A depiction, not a customer. No clinic is named here. The examples are the platform's own illustrations, marked example data on the home page. The WhatsApp version is illustrated, not captured from a real clinic: WhatsApp is live on Kav, but a clinic runs it on its own WhatsApp Business account, which Meta has to approve first.
What a patient asks, and what Kav answers
- "When is my next appointment?" Answered from the clinic's scheduling system: the day and time, the clinician, the clinic and the floor. Reschedule and Directions sit beside the answer.
- A front door the clinic builds with no code. A menu of Appointments, Billing and Test results, written as rows in the Studio's menu editor. A locked item is shown rather than hidden, with "sign in required" beside it, so nobody is surprised by the sign-in (the front door you build with no code).
- Health data under a stricter policy. Health-adjacent modules are held to the strictest sensitivity class before a single record is read. They are never sampled into analytics, and no summary is written if the conversation reaches a person (health questions under a stricter policy).
Where the money moves
The bold rows are the levers a clinic leans on most.
| Lever | What Kav does here | What you measure |
|---|---|---|
| Fewer repeat contacts | Appointment and location questions are answered from the scheduling system. | Reception calls and messages that are pure "when" or "where" lookups. |
| Faster when a person steps in | A handover carries the conversation, so reception does not start from "how can I help?". | Handling time on transferred contacts. |
| Out of hours, every channel | A patient gets appointment details at night, on the channel they already use. | Evening and weekend contacts, and missed calls that were really lookups. |
| Fewer errors and rework | Times and rooms come from the scheduling system through a template; the assistant does not type them. | Wrong-time and wrong-room complaints. |
| Revenue kept | Reschedule is one tap from the answer, so a patient who cannot attend moves the slot rather than not turning up. | No-shows and late cancellations, before and after. |
| Compliance and audit | Health data runs under the strictest policy by default, and every staff view of personal data is an audit record (the audit ledger nobody can edit). | Time spent preparing for a privacy review. |
Work out your own number
- Sample a week of front-desk contacts and count the pure lookups.
- Cost per contact = reception hours × loaded hourly cost ÷ contacts handled.
- Monthly value = lookups answered without a person × cost per contact, plus any no-show reduction you can measure in your own diary.
Work out your break-even from the published prices, or talk to us about connecting your scheduling system.