Research and practice

Visit confirmation is not a reminder. It’s a way to find out who won’t come while the slot can still be given to someone else.

What research and the practice of major booking platforms say: when to ask, whom to ask and what not to do, so clients don’t get annoyed and free slots don’t go to waste.

8.4% instead of 11.1%no-shows when the message named the cost of a missed appointment [10]
55–65% cheapera message works as well as a call [7]
−24%no-shows from one message a day ahead, vs −9% three days ahead (preprint) [5]
Chat with the salonexample
5 days before

Your time is booked: 20 November at 14:00. Are you planning to come?

The day before

We look forward to seeing you tomorrow at 14:00. Please confirm that you will come 🙌

2 hours before

We look forward to seeing you at 14:00 💛

The key point

The goal: hear “I can’t come” as early as possible

A cancellation a week ahead costs almost nothing: there’s time to give the slot to another client. A no-show on the day costs an hour of the specialist’s work. So the goal of every message is to hear “I can’t come” as early as possible and pass the freed slot straight to those who are waiting.

Cancelled 5 days ahead

The administrator offers the time to a client from the waitlist. The specialist’s day is full.

Didn’t show up

The specialist is waiting, the slot is empty, and there’s no one left to give it to.

How often to ask

How often to ask depends on when the client booked

The earlier the booking, the more often people don’t show up and the more messages are needed. Choose how many days before the visit the client booked.

  1. At bookingA message with the date, time and address. No question: the booking is fresh
  2. 2 hours before“We look forward to seeing you at 14:00”

The booking is fresh and plans haven’t had time to change, so there’s no need to ask.

This scheme is our conclusion from the research and practice below. The early question and the day-before confirmation are different: the first only finds out the client’s plans, the second confirms the visit. Anyone who tapped “I confirm” then gets only the “we look forward to seeing you” message 2 hours before.

What works

What works

  1. 1The client names the date and time themselves

    When patients wrote down the appointment date and time themselves, no-shows fell by 18%. Together with repeating it aloud and a poster in the waiting room about how many people arrive on time, they fell by 31.7%.

    Study NHS clinics, UK; before-and-after comparison [1]

  2. 2Booked far ahead — ask the first question earlier

    The time between booking and visit is the main predictor of a future no-show. At an eye clinic, 9.1% of patients booked 0–2 weeks ahead didn’t come, versus 38.3% of those booked six months ahead. When a children’s rehabilitation center cut its longest booking lead time from 12 to 4 weeks, last-minute missed visits fell by 8–16%.

    Study Appointments in hospitals and clinics, Iran and Canada; ophthalmology — a residents’ clinic [2][3][4]

  3. 3One message the day before works best

    One message a day ahead cut no-shows by 24%, one 3 days ahead by 9%. Two messages (3 days and 1 day ahead) cut them by 20%, so no better than one, but clients cancelled on their own more often. A review of SMS studies found no significant difference between 24, 48 and 72 hours. Conclusion: the day-before message is a must; an early one only where the booking is far ahead.

    Study 475,214 visits, Buenos Aires, WhatsApp; preprint, model on historical data [5][6]

  4. 4Message everyone, call those who don’t reply

    A message works as well as a call, and getting a client to show up costs 55–65% less. Calling clients at high risk of a no-show cut no-shows from 29.2% to 22.8%; in three trials, calls based on a risk model gave an average of minus 39%.

    Study Cochrane review and randomized trials [7][8][9]

  5. 5Name the specific cost of a no-show

    Only one sentence in the message was changed: it stated how much a missed appointment costs the hospital. No-shows were 8.4% instead of 11.1%, at no cost at all. A cost stated “in general terms” helped less — 9.9%. For a salon, it could be “your specialist is holding an hour for you”.

    Study Two randomized trials, about 20,000 patients, Barts Health, London [10]

  6. 6“Reschedule” in every message, and the reply updates the booking by itself

    The easier it is to cancel, the earlier people tell you. UCLH hospital lets patients cancel an appointment by replying to an SMS. Zenoti marks a booking as confirmed as soon as the client replies.

    Practice UCLH, Zenoti [11][12]

  7. 7A freed slot goes straight to those who are waiting

    Fresha offers a slot from the waitlist in three ways: one by one, to the most valuable client, or to everyone at once — whoever books first gets it.

    Practice Fresha: waitlist [13]

What doesn’t work

What doesn’t work

  1. 1Two questions in a row, or asking again someone who has already confirmed

    A question 2 days ahead and another one a day ahead add almost nothing to each other (see “What works”, 3). And a client who has already tapped “I confirm” sees that their answer wasn’t heard. They should get only “we look forward to seeing you” 2 hours before. An early “Are you planning to come?” a few days ahead and the day-before confirmation are not a repeat: they are two different questions with a gap between them.

  2. 2Asking about a far-off booking only the day before

    Plans changed weeks ago, and a slot freed up a day ahead is almost impossible to fill (see “What works”, 2).

  3. 3Calling everyone

    It’s expensive and works the same as a message (see “What works”, 4). Calls are for those who don’t reply.

  4. 4Booking two clients for the same time “in case of a no-show”

    The review did not prove any benefit of such double-booking based on no-show prediction, and for clients it means waiting and inconvenience.

    Study Review of studies [9]

  5. 5Cancellation by phone only

    Calling and explaining is awkward, so the client simply doesn’t come. Major platforms let clients cancel by replying to a message (see “What works”, 6).

  6. 6Starting with deposits and penalties

    First come messages, “Reschedule” and a waitlist. Deposits are for long treatments or after several no-shows.

  7. 7Early cancellation without a waitlist

    If the freed slot wasn’t offered to anyone, the early “I can’t come” achieved nothing.

Platform

What the platform does from this list

A question the day before with “I confirm”, “Reschedule” and “Cancel” buttons; “I confirm” and “Cancel” are recorded in alteg.io automatically, “Reschedule” starts a chat with the administrator in the same conversationLive
A reminder a few hours before the visit — to every client who is in the botLive
An early question a few days ahead, “Are you planning to come?”, with “I plan to come” and “Reschedule” buttonsPlanned
Waitlist and “Free slot”: when a booking is cancelled, the administrator sees who is waiting for this service and this timePlanned
Sources

Sources

Most of the research comes from medicine: outpatient clinics and hospitals. There are few studies of salons specifically, so alongside them is the practice of major booking platforms and chains. Every source is linked — check for yourself.

  1. StudyMartin S. J., Bassi S., Dunbar-Rees R. — Commitments, norms and custard creams: a social influence approach to reducing did not attends (DNAs). J R Soc Med, 2012 — NHS clinics: writing the date down themselves −18%, together with repeating it aloud and a poster −31.7%
  2. StudyAnisi S. et al. — Shiraz E-Medical Journal 19(8), 2018 — 148,077 bookings: the time between booking and visit is the main predictor of a no-show
  3. StudyMcMullen M. J., Netland P. A. — Lead time for appointment and the no-show rate in an ophthalmology clinic. Clinical Ophthalmology, 2015 — ophthalmology, 51,529 visits: 9.1% no-shows when booked 0–2 weeks ahead, 38.3% when booked six months ahead (residents’ clinic)
  4. StudyRavenscroft, Abouee Mehrizi, Wylie-Toal — MDM Policy & Practice 9(2), 2024 — children’s rehabilitation: booking lead time cut from 12 to 4 weeks — 8–16% fewer last-minute missed visits
  5. StudyEsteban et al. — medRxiv preprint, 2026 — preprint, 475,214 visits, WhatsApp: a day ahead −24%, three days ahead −9%, both −20%
  6. StudyGuy R. et al. — How effective are short message service reminders at increasing clinic attendance? A meta-analysis and systematic review. Health Services Research, 2012 — review: no difference found between 24, 48 and 72 hours for SMS
  7. StudyGurol-Urganci I. et al. — Mobile phone messaging reminders for attendance at healthcare appointments. Cochrane, 2013 — Cochrane review: a message works like a call, attendance is 55–65% cheaper
  8. StudyShah S. J. et al. — Journal of General Internal Medicine, 2016 — randomized trial: a call to the high-risk group, no-shows 29.2% → 22.8%
  9. StudyOikonomidi T. et al. — JAMIA, 2023 — review: calls based on a risk model −39% on average; benefit of double-booking not proven
  10. StudyHallsworth M. et al. — Stating appointment costs in SMS reminders reduces missed hospital appointments: findings from two randomised controlled trials. PLOS ONE, 2015 — two randomized trials: cost of a missed appointment in an SMS, no-shows 11.1% → 8.4%
  11. PracticeUCLH — SMS appointment reminder service — a hospital in London: SMS reminders and cancellation by reply
  12. PracticeZenoti — Automate appointment confirmations with 2-way SMS — two-way SMS confirmation: the reply marks the booking by itself
  13. PracticeFresha — Set up and manage your waitlist — waitlist: one by one, to the most valuable client, or to everyone at once

Visit confirmation in your salon

Leave your contact details and we’ll show you how the day-before question works with your alteg.io.

Or write to us directly:
founder@alteg-tech.com

What are you interested in (pick any)

By submitting the form you agree to the Privacy Policy.