Online scheduling and no-shows: what actually moves the numbers

The most common complaint a few weeks after online booking goes live is that the phone rings as much as before. It almost always has one cause: the patient books online, then calls to check it worked. The system gives out an appointment but not confidence — and confidence is what the call was for.

The second complaint is that the empty slots did not go away. That has a different cause, and it is the more expensive of the two.

Three numbers to measure before you start

Without these, you cannot tell afterwards whether anything improved — and you will end up arguing about impressions.

1. Your real consultation length. Not the number you would like; the number that happens. If visits take eighteen minutes and the calendar sells fifteen, you are an hour behind by lunchtime and the last patient of the morning waits for it.
2. The share of patients who arrive without an appointment. If it is forty per cent, a calendar that sells a hundred per cent of the day online breaks every single day.
3. Your no-show rate, by day and by appointment type. Measure it for a month before go-live, or you will never be able to say whether the new system helped.

Slot length is the whole schedule

Everything else in a schedule is downstream of the slot. Three adjustments do most of the work:

• Set the slot to the real length, then round up rather than down. An honestly slower calendar keeps its promises; an optimistically fast one breaks them by mid-morning and the delay accumulates all day.
• Use different lengths for different appointment types. A follow-up, a first consultation and a procedure are not the same appointment, and one average that fits none of them fits none of them.
• Leave two deliberate gaps a session. They are not idle time; they are where the day catches up with itself, and the cheapest way to stop one long visit cascading into a two-hour delay.

Walk-ins belong in the same calendar

The most common structural mistake is running two systems: a booked calendar in the software and a walk-in queue on paper at reception. The two then compete for one doctor, nobody can see the real wait, and the patient who booked three weeks ago watches someone who arrived twenty minutes ago go in first.

Put walk-ins into the same calendar as a queue with a position, so both kinds of patient occupy the one timeline the doctor is working through. Then reserve capacity explicitly: if walk-ins are a third of your day, sell two thirds online. Capacity you have not reserved is capacity you have oversold.

Reminders: timing, channel, and a reply

Reminders are the highest-return change available to most clinics, and are usually configured badly. Three rules cover most of the gain.

1. Send two, not one: a confirmation at the moment of booking, and a reminder twenty-four to forty-eight hours before. The confirmation is what stops the check-up phone call; the reminder is what stops the no-show.
2. Make the reminder answerable. A message that can be confirmed or cancelled with one tap converts a silent no-show into a cancellation with a day’s notice — and a day’s notice is a slot you can still sell.
3. Say what a missed appointment costs the clinic and how to release it. Most non-attendance is not indifference; it is a patient who assumed nobody would notice.

A cancellation only pays for itself if something is waiting to use it. A short waiting list, offered automatically to the next suitable patient the moment a slot opens, is what turns a day’s notice into a filled appointment rather than a gap.

What to watch after go-live

All five come out of data the booking system already holds, provided bookings, arrivals and visits are one record rather than three. That is the same argument as structured clinical data, applied to the front desk: an uncaptured number is a question you stop asking.

Mistakes that make the phone busier

• Requiring an account before the patient can see whether any slot exists. Show availability first, ask for details second.
• A form with fifteen fields. Everything but name, number and reason for visit can wait until reception.
• Publishing slots the doctor has not confirmed, then phoning to move them. Twice and patients go back to calling first.
• No visible cancellation route, which leaves the phone as the only way to cancel — so most patients do not.
• Bookings that do not reach the doctor’s screen instantly, so reception keeps a parallel list as insurance.

Frequently asked

How much does online booking actually reduce no-shows?

On its own it changes very little — the reduction comes from the reminders it makes possible. Clinics that add a confirmation at booking and an answerable reminder a day ahead typically see non-attendance fall by a third or more. Clinics that change the channel and send nothing see no change, which is why reminder settings matter more than the booking page.

Should we charge a fee for missed appointments?

It works better as a stated policy than as a collected charge. Clinics that publish a policy and make cancelling genuinely easy tend to get the behaviour without the confrontation. Charging patients who were never given an easy way to cancel mostly generates disputes at reception.

How far ahead should we let patients book?

Long horizons increase no-shows, because a lot changes in two months. Four to six weeks suits most outpatient work, with a waiting list for anything further out. If your specialty books months ahead, add a second reminder a week before.

Should we overbook to cover no-shows?

Only against a measured rate, and only on slot types where you have measured it. Overbooking a session with a ten per cent no-show rate turns one absent patient into two waiting ones on most days. A waiting list that fills cancellations gets the same occupancy without the crowded waiting room.

Will online booking replace the receptionist?

No, and a system sold on that basis usually creates work. What it changes is the mix: fewer calls to make and confirm simple appointments, more time for arrivals, payments and the patient standing at the desk. The measurable outcome is calls per booking, not headcount.