Solution blueprint · Healthcare · AI automation
An AI WhatsApp front desk for a multi-speciality clinic
Patients message the clinic on WhatsApp at any hour. An AI assistant answers in English, Hindi or Marathi, books and reschedules appointments against the doctors' real calendars, sends reminders and reports, and passes anything clinical to a person.
Design mockup with sample data. This is a solution blueprint by Infovion Technologies, not a client engagement.
- 24 × 7
- replies on the clinic's own WhatsApp number
- 3
- languages: English, Hindi, Marathi
- 6 – 8 weeks
- from kick-off to go-live
- ₹2.5 – 6 lakh
- typical build budget in India
The situation
The problem this solves
A busy clinic gets most of its enquiries on WhatsApp and by phone: timings, fees, which doctor, whether a slot is free tomorrow. Two receptionists answer them between patients at the counter, so messages wait for hours, calls go unanswered after 8 pm, and a share of booked patients simply do not turn up.
The clinic does not want a menu bot that makes people press 1, 2, 3. It wants patients to write the way they already do, get a correct answer from the clinic's own information, and reach a human immediately when the question is medical or the patient is upset.
The system
What it includes
Conversational booking
The assistant understands free text and voice notes, offers real slots from each doctor's calendar, books, reschedules and cancels, and confirms with a WhatsApp template message.
Reminders that reduce no-shows
A reminder the day before and two hours before, with one-tap confirm or reschedule. Freed slots go back into the calendar straight away.
Answers from the clinic's own information
Timings, fees, doctors, insurance panels, directions and preparation instructions are answered from documents the clinic controls, with the source logged for every answer.
Reports and prescriptions
Lab reports and prescriptions are sent as PDFs after the patient's mobile number and date of birth match the record.
Hand-over to staff
Clinical questions, complaints, emergencies and anything the assistant is unsure about go to a shared staff inbox with the full conversation. Staff reply from the same number.
Front-desk dashboard
Today's bookings by doctor, conversations waiting for a person, no-show rate, busiest hours and the questions patients ask most.
How it works
From input to outcome
- 01
Message arrives
WhatsApp Cloud API delivers the text or voice note to the server by webhook.
- 02
Understand
The language model classifies the intent and extracts doctor, date and time. Voice notes are transcribed first.
- 03
Check the rules
Emergency words, clinical questions and low confidence go straight to staff. The assistant never gives a diagnosis or dosage.
- 04
Act
Booking tools read and write the HMS calendar; answers are retrieved from the clinic's approved documents.
- 05
Reply and log
The reply goes back in the patient's language. Every message, tool call and hand-over is stored for audit.
What the design is built to achieve
- Routine questions answered in under a minute at any hour, without adding front-desk staff.
- Fewer no-shows through confirmed reminders, with cancelled slots released for other patients.
- No medical advice from the assistant: clinical and emergency messages always reach a person.
- Patient data kept in India, with consent recorded, in line with the DPDP Act, 2023.
Technology
WhatsApp Cloud API, Claude API, Node.js / NestJS, PostgreSQL with pgvector, Speech-to-text, React dashboard, HMS calendar API, AWS Mumbai
Budget and timeline
| Build, clients in India | ₹2.5 – 6 lakh |
|---|---|
| Build, clients outside India | $3,000 – 7,000 |
Timeline. 6 – 8 weeks, starting with a two-week proof of concept on the clinic's own questions.
Running cost. Model and WhatsApp message charges, estimated before the build; typically a few thousand rupees a month for a single clinic.
Indicative planning ranges for 2026. Every project receives its own fixed-price proposal after a first call. Source code and accounts belong to the client.
Request a fixed-price proposalFAQ
Frequently asked questions
Can an AI chatbot book doctor appointments on WhatsApp?
Yes. Through the official WhatsApp Business Platform, an assistant can read the doctors' calendars in the clinic's software, offer free slots, book, reschedule and cancel, and send confirmations and reminders. It needs an API or database connection to the appointment system.
Is it safe to use AI with patients?
It is, when the limits are built in. In this design the assistant never diagnoses or advises on medicines, sends emergency and clinical messages to staff at once, verifies identity before sharing reports, and logs every conversation.
How much does a WhatsApp chatbot for a clinic cost in India?
A menu-based bot costs roughly ₹40,000 to ₹1.2 lakh. An AI assistant connected to the appointment system, as described here, is typically ₹2.5 lakh to ₹6 lakh to build, plus Meta's per-message charges and model usage each month.
Does it work with our existing hospital software?
Usually. If the HMS has an API we use it; if not, we read and write appointments through a small connector on its database, agreed with the HMS vendor.
Is this a project you delivered for a client?
No. This page is a solution blueprint: a design we have worked through in detail, shown with sample data. Our delivered work is listed under case studies.
More blueprints
Other systems we have designed
Looking for delivered work? Read the case studies on our school ERP, jewellery software and website.
Planning a system like this?
Tell us how the work is done today. We reply within one business day with questions, an approach and a fixed-price proposal where the scope allows.