Clinic software for a polyclinic sharing one front desk

A polyclinic is not a bigger single doctor clinic. It is three or four lanes of patients in one waiting room, one desk registering all of them at different consult fees, one counter taking the money, and one owner who wants to know at day end who saw how many people. Two things decide whether software copes: what the price is charged per, because per practitioner pricing turns one bill into four, and whether the queue and the fee know that doctors differ. Feature lists answer neither.

The questions a shared front desk actually asks

1. Does the token restart at 1 for each doctor, or run as one series for the clinic?
2. A waiting patient is moved to another doctor. What happens to the number on the slip in their hand?
3. Does the fee field know the dermatologist charges Rs 500 and the physician Rs 300?
4. A patient saw one doctor on Monday and a second on Thursday, inside the revisit window. Free or paid?
5. Can a patient reach a Rs 500 lane having paid the Rs 300 fee at the shared counter?
6. A doctor does not turn up. How many actions empty that lane into a colleague's?
7. Can a doctor on a revenue share see the clinic's day book?

What it costs, and what the price is charged per

This is the part that decides a polyclinic's bill and the part the category hides. Every figure below was read off the vendor's own pricing page on 10 September 2026, not off a directory, and the column that matters is the third one.

Read that table for the unit column, not for the cheapest cell. At four doctors Elaior costs Rs 2,696 a month, which is more than every per clinic subscription above it: Practo Ray's Clinic Management is Rs 1,499 a month on a 1 year term and Rs 999 on the longest one, and KiViHealth's Rs 12,000 a year is about Rs 1,000 a month. The claim we defend is the narrower one, and it is about the unit. Elaior charges per clinic with a published per doctor add on, so a polyclinic can work out its own bill before it speaks to anybody, while Clinicea multiplies by practitioner and DocEngage by user account. Most of the field publishes nothing at all.

The directories show how little of this category publishes a number you can line up against another. On SoftwareSuggest, Practo Ray, MediXcel EMR and HIS, KareXpert, Clinic Master and MediSmart all sit as available on request, and where a starting price does appear the unit changes from row to row: Lybrate at Rs 799 a month per clinic, Praxify at Rs 120 a month per user, CrelioHealth LIMS at Rs 5,999 a month per centre. MocDoc names its product Clinic/Polyclinic Management Software and lists queue management among its features, publishes no price, and the clinic management route reached from its own home page returns a not found error.

One token series for the clinic, not one per doctor

In Elaior the token is a single clinic wide sequence that resets daily, enforced by a unique index on clinic, queue date and token number. The consequence at a polyclinic is direct: when the desk moves a waiting patient into another doctor's lane, the patient keeps their number. The slip in their hand stays valid and only the lane, and the position inside it, change. The older per lane renumbering was dropped, because it could collide under that same index.

All the lanes agree on who is next, because the board, each doctor's own list and the server's default pick share one ordering module. Rank runs called first, then critical, then an appointment whose slot has arrived, then walk ins and late arrivals, then an appointment parked because it arrived early, then booked but not arrived, then stepped out. Arriving within 15 minutes of a slot still counts as arriving on time.

Each doctor's own fee, and a revisit window that follows the doctor

A registration is priced by a three step cascade: the fee typed at the desk wins, then the doctor's own fee on their membership row, then the clinic default, and a free revisit overrides all three with zero. The doctor fee column is nullable, which is the whole trick. A doctor either carries their own number or falls through to the clinic's, so four doctors can hold four different fees and the desk has to remember none of them. Choosing a lane fills the fee field in from that doctor.

The free revisit window is measured against the patient's last visit with the SAME doctor, not their last visit to the clinic. The lookup filters on clinic, patient and doctor and skips deleted visits, and the window itself cascades patient, then doctor, then a clinic default of 7 days. So a patient who saw the physician on Monday and is sent to the paediatrician on Thursday is a fresh paid consult, and one doctor can be given a different window from another.

The shared counter, and the gate behind it

The billing worklist returns, for every patient sitting at billing, the token number, that visit's consult fee and the doctor's name, so the counter can see whose payment is whose without asking. If the doctor's user row has gone, the row still names somebody rather than nothing.

The pay first rule is enforced on the server, not in the browser, and there is no admin bypass. If the visit's fee is zero or unset the door opens, which is what a free revisit gets. Otherwise the code looks for a paid consult invoice on that visit and refuses without one, and that check sits on both admission paths: the desk sending a patient in, and the doctor starting the consult. Calling a token is not gated; admission is. The invoice it looks for is raised from that visit's own consult fee, and that fee came from the cascade above, which read it off that doctor.

When a doctor is called away or never arrives there is a one action move. It takes every entry in that lane that has not started a consult and appends it to another lane in the order it already had, each patient keeping their token, and it deliberately leaves a consult in progress alone. It is seated to the admin and the front desk, so a lane can be redistributed without a doctor signing in.

What the owner sees, and what the doctors cannot

There is one day book for the whole clinic, over any range, with the day end sheet as its single day detail: what was collected with refunds netted out, split by what it was paid for across consult, pharmacy, procedures and other; what went out, as dated expense rows each carrying a category; and the net of the two. The day book, the day end sheet and the owner dashboard are closed to every doctor seat in the authorisation table, not by a hidden menu item.

The waiting room screen with several lanes

The display board is built for more than one lane in one room: a panel per doctor, the grid splitting evenly by the number of lanes, each panel carrying that doctor's name and speciality with the token as the hero. A scope in the URL narrows it to one doctor full screen, so a polyclinic with two waiting areas can run a screen per lane. The wait estimate is that doctor's own running average of called to ended, and the board withholds it until they have 10 finished consults, because a wrong number a patient plans their morning around is worse than no number.

Four things we will not dress up

• The doctor dropdown at registration lists doctors by name only. Speciality is set in settings and shown on the waiting room board, not in that dropdown.
• The day book cannot split takings by payment method. Nothing in the app asks or records how a patient paid, so the money is one figure however it arrived.
• The monthly workbook for the clinic's accountant refuses to build unless the clinic timezone is Asia/Kolkata.
• Elaior has no paying customers today, so there is no polyclinic reference for you to call. Everything above is code, read on 11 September 2026, and not a roadmap.