Estimated reading time: 4 minutes
Part of Selling AI to Indian Hospitals, a field guide.
You can win the buyer, clear the data questions and run a converting pilot, and still stall at the last step: “how does this actually plug into what we already run?” An Indian hospital is not a blank slate. It has a system for appointments, a system for records, a WhatsApp number the front desk already uses, and a receptionist who will decide in a week whether your AI is a help or a headache. Integration is where deals get real, and where a lot of them quietly break.
At BeyondChats we build AI that talks to patients, which means we live at exactly these seams. Here’s a practical map of what you’re integrating with in an Indian hospital, and where the friction actually is.
The three seams that matter, in one line each:
- The HIS/HMIS: the hospital’s core system for appointments, patients and billing.
- WhatsApp: where Indian patients actually are, and the channel with its own rules.
- The human workflow: the receptionist and the doctor whose day you’re changing.
1. The HIS/HMIS: the system of record
Most hospitals run a Hospital Information System (HIS) or Hospital Management Information System (HMIS): the software that holds appointments, patient records, doctor schedules and billing. If your AI books an appointment or looks up a patient, it has to talk to that system, or it’s just a chatbot that creates work for someone to re-key.
The reality on the ground: there is no single standard. Some hospitals run well-known HIS products with real APIs; many run older, customised or homegrown systems with no API at all. Before you promise “we’ll integrate,” find out three things: what HIS they run, whether it exposes an API (or only a database, or nothing), and who owns that relationship: the hospital’s IT team or an external vendor who will need to be in the room. “We integrate with everything” is a promise you cannot keep; “we integrate with yours, and here’s how we handle the ones without an API” is one you can.
Where there’s no API, have a fallback that doesn’t require one: a lightweight sync, a structured handoff, or running alongside the HIS rather than inside it, so a missing integration doesn’t kill the deal.
2. WhatsApp: where the patients actually are
In India, the patient is on WhatsApp, not on your web widget. That makes the WhatsApp Business Platform the highest-value channel for patient-facing AI, and it comes with rules that catch newcomers out. You need a verified business account, message templates that Meta pre-approves for anything you initiate, and an understanding of the 24-hour session window (outside it, only approved templates can be sent). Pricing is per-conversation, not per-message, which changes the economics of reminders and follow-ups.
None of this is hard once you know it, but it is not optional, and it is not instant: template approval and account verification take time. Build that lead time into the pilot plan, or you’ll promise a go-live date you can’t hit for reasons that have nothing to do with your product.
3. The human workflow: the receptionist decides
The most under-estimated integration isn’t technical. It’s human. Your AI lands in the middle of a front desk that already has a way of doing things. If it dumps half-finished conversations on the receptionist, contradicts what the doctor told a patient, or needs constant babysitting, it will be quietly switched off no matter how good the demo was.
Design for the handoff: when the AI can’t help, it should pass a clean, summarised context to a human, not a wall of transcript. Give the front desk visibility and an off-switch. Make the doctor’s life easier, not noisier. The integration that decides renewal is the one between your product and the person who has to work next to it every day.
This is part of what I’m writing about here: selling and building AI for Indian healthcare, in public. If that’s your world too, here’s why I started writing, and the monthly letter below is where the numbers and the messier lessons go.
More in the field guide: integration questions surface in every pilot, and they’re tangled up with the data questions a hospital will ask. Or see all the guides. On the operations side, WhatsApp is also how you cut patient no-shows and keep an OPD queue moving.
Get the monthly letter
Building BeyondChats in public: the numbers, the decisions, and what broke. One email a month. No spam, unsubscribe anytime.