aiReceptionistaiReceptionist

How to Reduce No-Shows: 7 Data-Driven Measures and How to Automate Them

Team at aiReceptionistWritten with AI assistance21 min read

In summary

  • The average no-show rate reported in the scientific literature is 23%, with a huge range across sectors and countries (Dantas et al., Health Policy, 2018, review of 105 studies).
  • The two factors that most predict a no-show are how much time passes between booking and the appointment day, and having missed an appointment before. Both can be addressed through your scheduling setup.
  • Reminders work, but solid evidence is for SMS: they improve attendance with a relative risk of 1.14 (Cochrane, 2013). Email operates on the same mechanism, but the effect is not measured as well.
  • What really recovers money is not avoiding no-shows, but filling the gap: a waiting list that notifies as soon as someone cancels.
  • aiReceptionist, the AI receptionist from AIsToHire, automates email reminders, absence policies, waiting lists, and no-show reports. Does not send SMS or WhatsApp and does not charge for signals: it's good to know this before starting.

If you have a business that relies on scheduling —a dental clinic, an aesthetics center, a consultancy, a workshop, a physiotherapy office— you already know how painful an empty chair at eleven in the morning is. The client doesn’t notify, the gap can no longer be sold, and the cost of the premises and staff continues to accrue. This article is a practical guide to lowering that number: first the data on how much is actually missed and why, then seven concrete measures, and finally what part of all that can run on its own with aiReceptionist, the AI receptionist from AIsToHire and what part cannot.

Honesty warning from the start: here you will not read that a tool reduces no-shows by 80%. That figure does not exist in any serious study, and anyone who promises you that is selling. What does exist is a set of measures with measured effects and a couple of automations that save you from doing them manually every day.

Reception of a neighborhood clinic at sunset with an empty waiting chair and a laptop with the agenda open
Reception of a neighborhood clinic at sunset with an empty waiting chair and a laptop with the agenda open

How many people really miss an appointment and how much does it cost you

The average no-show rate in international literature is 23%, according to the systematic review by Dantas and colleagues published in Health Policy in 2018, which synthesized 105 studies from 1980 to 2016. But that average hides rates from 4% to figures above 40%: the African continent had a rate of 43.0% and Oceania 13.2%. Your number does not resemble anyone else's average; it must be measured in your own schedule.

These are the most useful published data to place you:

AreaNo-show rateSource and date
Average of 105 studies (1980–2016)23% (Africa 43.0%; Oceania 13.2%)Dantas et al., Health Policy 122(4):412-421, 2018
VA public hospital network (U.S.), ~1 million appointments per year18.8% on average; from 12.6% in Audiology to 25.7% in GastroenterologyBMC Health Services Research, 2016
Private ophthalmology consultation (Germany), 16,894 appointments1.8% if booked by the patient online versus 5.9% if booked by staff (p < 0.0001)Frontiers in Digital Health, 2025
University hospital (Germany), 81,173 appointments14.3% online versus 11.2% offline (p < 0.0001)Frontiers in Digital Health, 2025
Public primary care in Galicia, year 20243.86% (702,504 appointments)Sergas data reported by the press, January 2025
Public hospital consultations in Galicia, year 20245.75% (245,535 appointments)Sergas data reported by the press, January 2025

Pay attention to the two German rows, as they hold the most important lesson of the entire table: in the private consultation, allowing the patient to choose their slot reduced no-shows to less than a third; in the university hospital, where the online request was merely a request that later went through triage, the opposite occurred. Booking online does not reduce no-shows by itself. What reduces them is allowing the client to choose their time, see it confirmed, and feel that it is theirs.

The cost is not the price of the service, it is the gap that no one used

In the U.S. VA system, the average marginal cost of each no-show was estimated at 196 dollars in 2008 (BMC Health Services Research, 2016). That figure is not transferable to a neighborhood hair salon, but the calculation is:

Completed appointments per monthNo-show rateAverage service priceRevenue that evaporates
2505%35 €438 € per month
4008%60 €1,920 € per month
12015%90 €1,620 € per month

With a nuance that is important not to forget: you only lose that money if the gap remains empty. If you fill it with another person, the no-show costs you the inconvenience, not the revenue. That’s why measure 5 of this article (waiting list) usually pays off more than any other.

Comparison of the same schedule gap: on the left the empty consultation chair with cold light, on the right the same chair occupied by a client attended by a professional
Comparison of the same schedule gap: on the left the empty consultation chair with cold light, on the right the same chair occupied by a client attended by a professional

The two factors that most predict a no-show are lead time and history

The same review from Health Policy (2018) identifies a long wait time between booking and the appointment and having a previous history of absences as the main determinants of no-shows. The other factors (young age, low socioeconomic status, distance to the appointment) matter, but they are out of your control. The first two are not.

FactorWhat the evidence saysCan you act?
Lead time between booking and appointmentMain determinant: the further in advance it is booked, the more likely it is to be missed (Dantas et al., 2018)Yes. Limit how many days in advance can be booked
Client's history of absencesMain determinant: those who missed before will miss again (Dantas et al., 2018)Yes. You need to record who missed to be able to use it
Day of the weekMondays have the highest concentration of absences (BMC Health Serv Res, 2016)Partially. Book Mondays for clients with a good history
Type of serviceSpecialties have more no-shows than general consultations (BMC Health Serv Res, 2016)Partially. Check your report by type of appointment
Who made the bookingAllowing the client to choose their slot reduced no-shows from 5.9% to 1.8% in private consultations (Frontiers in Digital Health, 2025)Yes. Give real autonomy, not a request to confirm
Age, income, distanceAssociated with no-shows (Dantas et al., 2018)No. And using them to discriminate would be a bad idea

The 7 measures that reduce absences, ordered by effort

If you are only going to do two things, do number 2 (reminders) and number 5 (waiting list): the first prevents some absences and the second turns the ones you don’t prevent into money. All seven together make the complete program, and none require changing the price of your service.

#MeasureEffortWhat factor it attacksIs it automated?
1Shorten the booking window1 minuteAdvance noticeYes
2Two or three staggered reminders10 minutesForgettingYes
3Make canceling easier than not showing up5 minutesFrictionYes
4Written and applied absence policy5 minutesHistoryYes
5Waiting list with automatic notification5 minutesEmpty slotYes
6Pre-appointment form15 minutesCommitmentYes
7Mark every day who came and who didn’t1 minute a dayMeasurementNo, it’s manual

1. Shorten the booking window

This is the cheapest measure and it attacks the number one factor. If today someone can book an appointment for three months from now, you are giving away absences. In aiReceptionist, this is the "booking window" field in Settings → Schedule: by default, it is 21 days ahead and can be adjusted between 1 and 365. Next to it are the minimum advance notice (default 2 hours: prevents someone from booking for ten minutes from now) and the granularity of the slots.

A practical tip: if your business has seasonal treatments or biannual check-ups, don’t just drop the window to 15 days. Shorten it for short services and leave the long window only where the appointment is truly planned.

2. Send two or three staggered reminders

The best available evidence is the Cochrane review by Gurol-Urganci and colleagues (2013, CD007458.pub3): SMS reminders improved attendance compared to not sending anything, with a relative risk of 1.14 (95% CI 1.03–1.26), moderate quality evidence from 7 studies and 5,841 participants. Adding SMS to a postal letter also helped (RR 1.10; 95% CI 1.02–1.19), and the cost per message turned out to be between 55% and 65% cheaper than a call.

Translated to your schedule: two notices, one with a day's notice and another on the same day. In aiReceptionist, they are configured in Settings → Reminders, choosing from the steps of 48 h, 24 h, 4 h, 2 h, and 1 h before; the default are 24 h and 2 h, and you need to activate the switch because they come deactivated. The subject and text are editable with variables: {name}, {date}, {time}, {type}, {business}, {professional}, {location}, and {management_link}.

Two honest limits before you set it up:

  • Reminders are sent by email. SMS and WhatsApp channels are planned in the system, but today they are marked as omitted because there is no connected messaging provider. The evidence I just cited is from SMS; for email, the mechanism is the same, but don’t assume the same effect.
  • Without email, there is no reminder. If the client only left their phone number, the sending is omitted. That’s why the email is marked as a mandatory field by default in the client data configuration: if you remove it, you’ll have no reminders for that person.
Nightstand with a mobile phone lit up by a notification received the previous afternoon, next to an analog alarm clock
Nightstand with a mobile phone lit up by a notification received the previous afternoon, next to an analog alarm clock

3. Make canceling easier than not showing up

Many people don’t notify simply because notifying is uncomfortable: they have to call during business hours, give explanations, and endure silence on the other end. If canceling costs two written sentences at eleven at night, many of those people will cancel — and leave you the slot in time.

In aiReceptionist, each reminder email includes a "Manage my appointment" button that opens the business's public chat. From there, the client writes that they cannot go, and the AI cancels or reschedules the appointment, always within two rules that you set in Settings → Schedule:

  • Minimum hours to cancel or move (default 2, up to 168). If the client asks to cancel later than that timeframe, the AI does not do it on its own: it passes the conversation to you with a summary.
  • Maximum number of changes per client (default 3, up to 10). Once the limit is exceeded, the conversation also goes to your inbox.

It’s not a self-service portal with buttons: it’s a conversation. For the client, the practical difference is that they don’t have to look for any management link or remember a password.

4. Write your absence policy and let it apply itself

A policy that lives in the owner’s head is not a policy. In Settings → Reminders → Absences, there are four options, and it’s important to understand exactly what each one does:

PolicyWhat it really does
NoneNothing happens. The history continues to be recorded
Notify the clientIf the client has missed before, the AI gently reminds them when booking
Request depositThe AI notifies that the business will ask for a deposit. There is no charge: no payment gateway exists. The amount and whether it is paid are noted by hand
Block after N absencesAfter N absences, that contact cannot book online: neither through chat nor the public booking page. They can still book by calling you

The count is done by phone or email of the contact (or by their client record if you use the CRM), and the default value for N is 3.

5. Maintain a waiting list and notify as soon as a slot opens up

This is the measure that turns an absence into a non-absence, and that’s why it is often the most profitable. When someone wants an appointment that you don’t have, the AI offers to put them on the waiting list. When an appointment is canceled —regardless of whether the client cancels it from the chat or you do it from the schedule—, the system looks for the first person on the waiting list that fits that type of appointment and that date, sends them an email with the link to book, and notifies you.

What it does not do: book for that person. The slot is offered to them, but whoever grabs it first gets it. It’s deliberate, because assigning an appointment to someone who hasn’t confirmed it is the best way to generate another absence.

A professional from an aesthetics center receiving a client who arrives to take a slot that has just opened up, with a tablet on the counter showing the schedule
A professional from an aesthetics center receiving a client who arrives to take a slot that has just opened up, with a tablet on the counter showing the schedule

6. Request the pre-appointment form before the appointment

When someone spends three minutes filling out a form for their appointment, they have invested something in it. It’s a reasonable commitment mechanism and also saves you time on the day of the visit; that said, I don’t know of any trial that quantifies its effect on absences, so treat it as a good practice, not as a figure.

In aiReceptionist, the pre-appointment form is activated by type of appointment (Settings → Appointment Types), with the fields you define. The client receives a unique link, and if they haven’t filled it out when the reminder comes, the email automatically adds the line to complete it. You can also resend it manually from the appointment record.

7. Mark every day who came and who didn’t

Without this step, the previous six measures are faith. The absence rate can only be calculated if someone distinguishes between "came" and "did not come", and that cannot be guessed by a program.

After the margin you set (0, 1, 2, or 4 hours after the appointment time; default 2), the appointments still marked as confirmed appear on the home panel under "appointments to review", with two buttons: Came and Did not come. That’s two taps per appointment. In return, you get three things: the real rate, the automatic notification of rescheduling to the client who missed (if you leave it activated), and the risk signal for that person's upcoming appointments.

That risk signal is calculated with the contact's history: it starts from a low base, increases by 30 points for each previous absence, decreases by 20 for completed appointments, increases by 20 if there is neither phone nor email to notify, and classifies the appointment as low risk (less than 35), medium (35 to 59), or high (60 or more). It’s not a sophisticated prediction: it’s your history, placed where you can see it.

How to measure if it's working: three numbers and a report

Give it between 60 and 90 days before judging anything, because with few appointments per month the rate jumps due to statistical noise. After that time, look at just three numbers and compare them with the same previous period.

NumberWhere it is in aiReceptionistHow it is calculatedAlarm signal
No-show rateMetrics (periods of 7, 30, 90 days or 1 year)Absences Ă· (completed + absences)Above 15% is marked in red
Distribution by day of the weekMetrics → No-shows tabCount of absences by dayA day that concentrates double than the rest
Repeat clientsMetrics → No-shows tabContacts with more than one absence in the periodThat the list grows month by month

The no-show report also shows the list of missed appointments for the period, so you can review specific cases instead of discussing averages. And the conversation report, on the same screen, can be exported to CSV if you want to cross it with your accounting.

What part of all this aiReceptionist does and what it does not

aiReceptionist chats with your clients, books in the real gaps of your schedule, and handles reminders, the waiting list, the absence policy, and the report. What it does not do, today, is send SMS or WhatsApp, charge deposits, or write in your Google Calendar. This is the complete table, contrasted against what the system actually executes:

What you need to reduce absencesDoes it do it?Detail
Automatic reminder before the appointmentYes, by emailSteps of 48, 24, 4, 2, and 1 hour before; subject and text editable with variables
Reminder by SMS or WhatsAppNoPlanned in the system, but without a connected provider: those messages are marked as missed
Cancel or move the appointment without callingYes, via chatWith minimum notice and limit of changes; out of time it goes to a person
Charge a deposit or feeNoYou can note amount and payment status; there is no gateway or automatic charge
Notify the client who missed to rescheduleYes, by emailTriggered when marking "Did not come", if you leave it activated
Block those who repeatedly missYesBlocks online booking (chat and public page); does not prevent them from calling you
Waiting list with notification when a slot opensYes, by emailNotifies the first that fits by type and date; does not book for them
Follow-up after the appointment and request for reviewYesAt 1, 3, or 24 hours, with your review link
Pre-sheet by type of appointmentYesOwn link, reminded in the email if pending
Absence reportYesRate, distribution by day, repeaters, and list of appointments
View the schedule in Google, Outlook, or AppleRead-onlySubscribable ICS feed; does not write appointments in your external calendar
Browser push notificationsNoNotifications to the owner arrive by email and within the app itself

How to set it up in about fifteen minutes

Order matters: without appointment types and without a schedule there are no gaps, and without gaps, there is nothing to remember. These are the seven steps, with the exact route of each screen.

  1. Settings → Schedule. Time zone, time slots for each day, duration of the gaps, and, at the bottom, the three rules that matter here: booking window (starts at 21 days), minimum notice (2 hours), and minimum hours to cancel (starts at 24 if your service is expensive per slot; 2 if it is short).
  2. Settings → Appointment types. One type per service, with its actual duration. Activate the pre-sheet for those that need it.
  3. Settings → Reminders. Turn on "Send reminders" and add the 24 h and 2 h steps. Rewrite the text in your words and leave the variable {enlace_gestion} inside: it allows them to cancel instead of disappearing.
  4. Same screen, "After the appointment" block. Follow-up at 3 or 24 hours with your review link. If you don't want it to go out before you mark the appointment, leave "send even if you haven't reviewed the appointment" deactivated.
  5. Same screen, "Absences" block. Mark to review from 2 hours, rescheduling message activated, and, if needed, the blocking policy after N absences.
  6. Settings → Waiting list and Settings → Share. Check the waiting list and copy your public link to paste it on the website, in the Instagram bio, and in your Google listing.
  7. Every morning, home screen. Check "appointments to review" and mark Came / Did not come. It is the only manual step, and it is what makes everything else measurable.

Where is the limit if your business is healthcare or regulated

aIReceptionist is a reception and scheduling tool: it informs about your services, collects data, and books appointments. It does not perform clinical triage, does not provide medical, legal, or tax advice, and does not make any decisions about anyone's treatment. When sensitive words appear —urgency, complaint, claim, lawyer, report, among others that you can expand— the conversation is flagged and passed to a person before the AI continues.

Three more things to clarify if you work in health, law, or finance:

  • The data from the pre-appointment form is yours, and so is the responsibility. If you request health information in a pre-appointment form, you are the one who responds as the data controller. The app provides configurable consent text, automatic retention period, and export and deletion of conversation data at the request of the interested party, but the legal basis and the information you request are decided by you.
  • Blocking someone for not showing up is your decision, not a recommendation. In a public or contracted healthcare service, there may be rules that prevent it. Consult with your advisor before activating it.
  • Nothing in this article is professional advice. The cited figures come from the linked studies and describe specific populations, not your business.

Frequently Asked Questions

What is a normal no-show rate in an appointment business?

It depends so much on the sector that the global average of 23% (Dantas et al., Health Policy, 2018) serves little as a reference. In Spanish public healthcare, the published figures for 2024 in Galicia were 3.86% in primary care and 5.75% in hospital consultations. What is useful is not to compare yourself with anyone, but to measure your rate over 90 days and monitor if it goes up or down.

How many reminders should I send and when?

Two are usually sufficient: one with a 24 or 48-hour notice, so that the client has time to cancel if they cannot make it, and another on the same day, so they do not forget. A third reminder an hour before only makes sense if your client travels little and can react. More than three reminders for an appointment start to seem like spam.

Do email reminders work the same as SMS?

It cannot be said that they do. The moderate-quality evidence that exists (Cochrane review, 2013) measured SMS, with a relative risk of attendance of 1.14. The mechanism of email is the same —to remind and provide the opportunity to cancel— but its open rate is different, and there is no equivalent number to cite. Today, aIReceptionist sends via email; SMS and WhatsApp channels will appear when there is a connected messaging provider.

Can I charge a deposit to avoid no-shows?

With aIReceptionist no, at least not automatically: you can set a deposit amount per type of appointment, have the AI communicate it to the client, and note if it is paid, but there is no payment gateway or card charge. If charging a deposit is essential for your business model, today you will have to manage it outside the app.

Is it a good idea to block a client who has missed appointments several times?

It is a business decision with a cost. Blocking online booking prevents the problem from recurring, but you also lose clients who missed for a valid reason. A middle ground that works well: leave the policy at "notify the client," which tactfully reminds them of their history when booking, and reserve blocking for those who accumulate three no-shows without notifying.

What do I do with the gap left by someone who cancels at the last minute?

Offer it to someone before it cools down. That’s what the waiting list is for: when an appointment is canceled, the system notifies by email the first person on the waiting list that fits that type of appointment and that date, and sends you a notice. Whoever responds first gets it; no one books on behalf of anyone.

Does aIReceptionist notify clients via WhatsApp?

No. Client notifications —reminders, follow-ups, absence messages, waiting list— are sent via email. SMS and WhatsApp sends are contemplated in the system but are marked as skipped until there is a connected messaging provider. If someone tells you otherwise about this app, it is not true.

How long does it take to see if my no-show rate has decreased?

Between 60 and 90 days, and with a reasonable minimum of appointments. If you handle 40 appointments a month, a no-show more or less moves the rate two and a half points: at that scale, monthly variations are noise. Compare complete quarters and also look at the distribution by day of the week, which usually provides more actionable insights than the average.

In a nutshell

No-shows are not eliminated, they are managed: shorten the notice period, remind twice, make cancellation easy, write your policy, have a waiting list, and measure who showed up. aIReceptionist is the virtual receptionist with AI from AIsToHire that handles chat, books in the real gaps of your schedule, and keeps email reminders, the waiting list, no-show policy, and no-show reports running automatically; it is available at aiReceptionist — AI receptionist that answers and books appointments What it does not do —SMS, WhatsApp, charging deposits— is stated above plainly, because choosing a tool knowing its limits is cheaper than discovering them along the way.

Sources

  • Dantas LF, Fleck JL, Cyrino Oliveira FL, Hamacher S. No-shows in appointment scheduling – a systematic literature review. Health Policy 122(4):412-421, April 2018.
  • 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, CD007458.pub3, 2013.
  • Prevalence, predictors and economic consequences of no-shows. BMC Health Services Research, 2016.
  • Efficient patient care in the digital age: impact of online appointment scheduling in a medical practice and a university hospital on the no-show rate. Frontiers in Digital Health, 2025.
  • Absenteeism data from the Servizo Galego de SaĂșde corresponding to 2024, collected by the press in January 2025.

Set up your reception with aiReceptionist: email reminders, waitlist, and no-show policy running on their own from day one. Create your free account and have your schedule set up in fifteen minutes.

Set up your reception with aiReceptionist: email reminders, waitlist, and no-show policy running on their own from day one. Create your free account and have your schedule set up in fifteen minutes.

↑ Back to top