How to cut no-shows, and what the evidence actually shows

A single no-show can cost a full day's rate. Over a month the empty chairs add up to real money, and they are almost always preventable. The fix is not chasing people harder. It is a quiet, consistent system that does the chasing for you.

Before any of that, one piece of honesty, because our own version of this article used to fail it. Booking software is sold with a familiar line: automate your reminders and no-shows drop by half. We used to print that sentence too, and we could not point to anything behind it. So here is what the research actually found, what it does not say, and how to work out the number that applies to your own calendar rather than someone's marketing page.

What the trials found

The strongest evidence available is a Cochrane systematic review of mobile phone messaging reminders for appointment attendance (Gurol-Urganci and colleagues, 2013, review CD007458), covering four randomised controlled trials and 3,547 participants.

Two results matter here. Against no reminder at all, text message reminders improved attendance with a risk ratio of 1.10, 95% confidence interval 1.03 to 1.17, on moderate-quality evidence. And against a staff member ringing each person, text reminders performed the same, risk ratio 0.99, 95% confidence interval 0.95 to 1.03, at a lower cost per attendance.

Read those two numbers together and they say something more useful than "cut no-shows in half". They say a text is as good as a phone call, and nobody has to make the calls.

Two caveats you should hold onto. First, these are healthcare appointments, not haircuts or tattoo sittings, and a dental check-up is not a Saturday colour appointment. Treat the direction as solid and the exact size as indicative. Second, moderate-quality evidence with a confidence interval starting at 1.03 is a real effect but a modest one, not a transformation.

Turning a 10% lift into your own number

Here is where the "half" claim comes from, and why it is not simply a lie, only sloppy. A lift in attendance and a fall in no-shows are different fractions of different denominators.

Suppose 20 out of every 100 booked clients currently do not turn up, so your attendance rate is 80%. Apply the 10% relative improvement from the review and attendance goes to about 88%. Your no-show rate has moved from 20% to about 12%. Expressed as a lift in attendance that is 10%. Expressed as a cut in no-shows it is closer to 40%.

Now run the same arithmetic on a calendar that already behaves. If only 5 in 100 miss, attendance goes from 95% to about 100% at best, and the absolute gain is five appointments, not forty per cent of anything meaningful.

The lesson is not that reminders work better for messier calendars, though that happens to be true. It is that the only percentage worth quoting is your own. Which means the first step is not switching anything on. It is measuring.

Measure your baseline before you change anything

Pick a month that has already finished, so nothing about it can still move. Count two numbers: appointments that were scheduled, and appointments where the client simply did not arrive. Cancellations with notice are a different problem with a different fix, so keep them out of this count.

That fraction is your baseline. Write it down somewhere you will find it again in sixty days. Without it, every claim you read, including ours, is untestable, and you will end up judging a reminder system by how the last fortnight felt.

Marking a missed appointment as a no-show rather than quietly deleting it is what makes this possible at all. In Skedful the status stays on the client record, so your rate over a month is a number you can look at instead of a feeling, and the assistant can report it back to you when you ask.

Why bookings get missed

Most no-shows are not flaky clients. They are forgotten bookings. An appointment made three weeks ago slips off the radar, and the further ahead it was booked, the more likely that is. The job is not to punish. It is to remind, at the right moment, in a way that is easy to act on.

That framing matters because it changes what you build. If you believe clients are careless, you reach for penalties. If you accept that memory is the problem, you reach for timing.

The three-touch reminder rhythm

1. At booking: an instant confirmation with the date, time and address. This is the touch that makes the appointment feel real, and it is also the one people screenshot.
2. Around 48 hours before: a reminder with a one-tap confirm link. Two days is far enough out that someone who cannot make it still has time to say so, and near enough that the date is inside their planning horizon.
3. A couple of hours before: a short "see you soon" with directions.

That is it. No spam, no guilt trips, just the right nudge at the right time.

Two things to resist. Do not add a fourth and fifth message: past a point you are training people to ignore you, and the review above found no benefit worth that risk. And do not write the reminder as a warning. A line about your cancellation policy belongs at booking, not two hours before the appointment.

Send it where the person actually reads

A reminder that lands in a channel someone ignores is not a reminder. This is worth a thought rather than a default: plenty of clients read WhatsApp within minutes and let SMS pile up unopened, and for some the opposite is true.

Skedful sends reminders over WhatsApp where the client is reachable there and falls back to SMS otherwise, with Israeli numbers going through an Israeli SMS route rather than an international one, because delivery rates and sender reputation differ by country. You do not configure any of that. The point of mentioning it is that channel choice is part of whether the message gets read, and it is the part most software leaves to chance.

Make confirming effortless

A reminder only works if acting on it takes one tap. Keep the confirm flow password-free. A phone number and a one-time code is enough:

// pseudo: a reminder that asks for nothing but a tap
sendReminder({
  to: client.phone,
  when: '48h-before',
  confirmUrl: shortLink(appointment.id),
});

When confirming is frictionless, more people do it, and the ones who cannot make it tell you early. That is the quiet win in the whole system: an early cancellation is not a loss, it is an opening.

The other half of the fix is the waitlist

Reminders reduce the number of people who forget. They do nothing about the slot that has just come free at four o'clock on a Friday. That is what a waitlist is for.

If a cancelled slot is offered automatically to the next person waiting, in the order they joined, then most late cancellations cost you a message rather than a chair. Pair the two and the arithmetic changes again: fewer people forget, and the ones who do are quietly replaced.

This is also the reason the reminder should invite an early answer rather than scold. A client who cancels at 48 hours hands you a slot you can still sell. A client who does not want to face a telling-off simply goes quiet, and you find out at the appointment time, when the slot is worthless.

What to do about the ones who still miss

Some people will still not turn up, and a system that pretends otherwise is not a system.

• Record it. A missed appointment marked as a no-show is on the record. Three in a row is a conversation, not a mystery.
• Watch the pattern, not the incident. Look at your rate by day and by time of day over a month. A single Friday evening tells you nothing, and a month of Friday evenings tells you plenty.
• Decide about instant confirmation. Letting bookings confirm themselves is the right default for most businesses, because friction costs you more real bookings than it prevents fake ones. If a particular service is expensive in chair time, keep manual approval for that service alone.
• Cap how far ahead people can book. A booking made four months out is a booking made by a different version of that person. A booking window keeps the calendar inside the horizon people can actually plan.

What we are not going to claim

We are not going to tell you that reminders will halve your no-shows, because the best available evidence does not say that, and because the honest answer depends on a number only you have. What the evidence does support is narrower and still worth having: an automated message performs as well as somebody phoning around, it costs less, and it reliably moves attendance in the right direction.

If your calendar is at 20% missed, the arithmetic above suggests a meaningful dent. If you are at 5%, the fix is worth having and will not change your life. Either way you will know, because you wrote your baseline down.

Want this running for your business? Start a free trial and you will be live in under ten minutes.

Source

Gurol-Urganci I, de Jongh T, Vodopivec-Jamsek V, Atun R, Car J. Mobile phone messaging reminders for attendance at healthcare appointments. Cochrane Database of Systematic Reviews 2013, Issue 12, Art. No. CD007458.