How to Increase Show Rate on Booked Sales Calls

Show rate = held calls ÷ calls booked in the same window. Raise it in two moves: fix the denominator so the number is real, then replace your single reminder with four touches across two channels, one of which asks for a reply. In a randomised trial of 54,066 patients, two reminders beat one — 4.4% missed versus 5.8%.

At a glance

• The metric: held calls ÷ calls booked, same fixed window. Three defensible denominators exist, 19 points apart on the same calendar.
• The floor: the 4–2–1 cadence rule — four touches, two channels, at least one reply request. One reminder is a notification, not a cadence.
• The best controlled evidence: two reminders (3 days and 1 day out) cut missed appointments to 4.4%, against 5.3–5.8% for one, across 54,066 randomised patients.
• The booking window: across 51,529 appointments, no-shows ran 9.1% at a 0–2 week lead time and 38.3% at six months, same clinic.
• Our own calendar: 10 of 27 bookings (37%) completed outside 9am–5pm Mon–Fri. Small sample, disclosed below — it decides whether touch one can be human.

How is show rate actually calculated?

Show rate is held calls divided by calls booked, both counted in the same fixed window. The arithmetic is trivial; the denominator is not, and that is where almost every reported show rate goes wrong.

Take one calendar month: 100 calls booked, 9 cancelled with notice, 13 moved to a later date, 12 never arrived and said nothing, 66 held in their original slot. Three people can compute three different show rates from those same rows, and all three are defensible:

Report booked-to-held: it is the only version of the number that cannot be improved by moving appointments around. Publish the other two beside it for diagnostics, but manage on 66%.

You also need a written rule for partial attendance, because calendars do not have one. Ours: a call counts as held only if the prospect joins within 10 minutes of the booked start and stays long enough to complete qualification. A four-minute “sorry, bad time” is a no-show wearing a green tick. For the difference between a no-show, a cancellation and a reschedule — three events, three fixes — see our page on no-show recovery rate, and its companion on cancelled appointment recovery. This page owns the calls that were never missed in the first place.

Why does one reminder not move my show rate?

Because a single reminder tests memory, and memory is not usually the failure. The best controlled evidence on reminder quantity is a three-armed randomised trial across 25 primary care clinics, published in The American Journal of Managed Care in 2018: among 54,066 randomised patients, those who received reminders both 3 days and 1 day before their visit missed 4.4% of appointments, against 5.8% for a 3-day reminder alone and 5.3% for a 1-day reminder alone. In the high-risk subgroup the gap was wider — 20.5% missed with two reminders, against 25.0% and 24.2% with one.

Two things transfer from that clinical setting to a sales calendar. The second touch is worth roughly a quarter of the remaining miss rate and costs nothing but a scheduled send; and its effect is largest on the bookings you were most likely to lose — on a sales calendar, cold and paid traffic rather than inbound demo requests.

One reminder is a notification. A cadence is four touches, two channels and at least one request for a reply — the 4–2–1 rule. The reply request is the load-bearing part: a “yes” is a commitment, and a silence is an early warning you can act on hours before the slot.

The reminder cadence: what to send, when, on which channel

The spec, not a suggestion. The last column is the honest per-booking cost of running it by hand.

Two content rules matter as much as the schedule. Put the time in their timezone — a timezone error reads to the prospect as your mistake and to your dashboard as their no-show. And make rescheduling one tap in every message: a reschedule is live pipeline, a ghost is not.

How far ahead should I let people book?

The booking window is the lever most teams never touch, and the evidence is unusually clean. A cross-sectional study of 51,529 appointments over 12 months in an outpatient ophthalmology clinic found no-shows rose steadily with lead time: at a 0–2 week lead time, 9.1% in the resident clinic and 2.4% in the faculty clinic; at six months, 38.3% and 6.9%. The authors’ model predicted roughly a 2% reduction in no-shows for every 10% more patients booked within two weeks.

That is a clinic, not a sales team, and the mechanism transfers rather than the magnitude: intent has a half-life, and every day between booking and meeting spends some of it. Default your scheduler to the earliest real availability, and treat a call booked more than seven days out as one that needs an extra touch, not one you can trust.

What we measured on our own calendar

Across 27 bookings completed on leadsnow.ai’s own booking pages between 29 July and 10 September 2026, 10 of them — 37% — were completed outside 9am to 5pm, Monday to Friday, Australian Eastern Time. Three landed between midnight and 3am, one on a Sunday evening, and the busiest single weekday was Friday.

Caveat, stated plainly: n = 27 is a small sample from one site over six weeks of our own inbound traffic. It is not a benchmark and we would not defend it as one. It is a reason to run the same query against your own booking timestamps before you decide who sends touch one.

Because that is the decision it settles. If your confirmation email and SMS go out when somebody next opens a laptop, then on numbers like ours more than a third of bookings sit unconfirmed for hours — the hours in which the decision to attend is most easily reversed. The first two touches must fire without a human, or they do not fire for the bookings that need them most.

Why is a ten-point show-rate gain worth more than it looks?

Because show rate sits in the middle of a chain of multiplications, and everything downstream of it inherits the gain. Run it end to end on 1,000 leads a month:

Four gains of 15% produce 1.15⁴ = 1.75× the deals — a 75% increase from four changes none of which impresses on its own. Move show rate alone from 55% to 63.25% and you get 9.5 deals instead of 8.25: a 15% lift for one project, and the only lever here that costs no extra media spend, because the appointment is already paid for.

The honest wrinkle: 1.75× is arithmetic, not a forecast, and the stages are not independent. Tighter qualification raises show rate and lowers set rate; a shorter booking window raises show rate and reduces how many people find a slot at all. Anyone quoting a large multiple built from stacked funnel gains is multiplying numbers that overlap. Treat compounding as a ceiling, and re-measure each stage after you change it — the stage-by-stage economics are laid out in our sales pipeline stages cluster.

What does running this cadence actually cost?

Add the last column of the cadence table: roughly nine minutes of human attention per booking before anyone replies to anything. At 100 bookings a month that is about 15 hours, a third of it outside business hours on timestamps like ours. The hours are not the hard part — the replies are. Someone must act on “can we move it to Friday?” within minutes, on a Sunday night, or the reply request is worse than sending nothing.

The middle row assumes a two-way SMS sender on a compliant number, a scheduler that emits booking and cancellation events you can trigger on, and one person who owns the weekly numbers. Automating that chain around the clock is what an AI appointment setting system is for. For the wider set of levers — message psychology, qualification tightness — our sales appointment show rate playbook covers the ground this page does not.

Frequently asked questions

How do I calculate my show rate?

Held calls divided by calls booked, over the same fixed window — a calendar month is the usual choice. Do not subtract cancellations or reschedules from the denominator. On a month with 100 bookings and 66 held calls your show rate is 66%, even though excluding 9 cancellations and 13 reschedules would let you report 84.6% from identical data.

Is one reminder enough?

No, and this is one of the better-tested questions in the whole funnel. A three-armed randomised trial of 54,066 patients published in The American Journal of Managed Care found that reminders sent both 3 days and 1 day before a visit produced a 4.4% missed-appointment rate, against 5.8% for a 3-day reminder alone and 5.3% for a 1-day reminder alone. Two automated reminders beat one, and patient satisfaction was unaffected.

How far in advance should I let people book?

As little as your capacity allows. In a study of 51,529 outpatient ophthalmology appointments, no-shows averaged 9.1% at a 0–2 week lead time and 38.3% at six months in the same clinic. Different setting, same mechanism: the further out the booking, the more competing priorities crowd out the reason it was made.

Should I count a reschedule as a no-show?

No — a reschedule moved before the start time is still live pipeline, so it belongs in neither the numerator nor the no-show count. But do not remove it from the denominator either, or your show rate improves every time a meeting is pushed. Track reschedules on their own line: a rising reschedule rate is a booking-window problem showing up early.

My show rate looks healthy but my closers say half the calendar is empty. Why?

Almost always a denominator or status-hygiene problem, not a disagreement about reality. Check three things in order: whether cancellations and reschedules are being excluded from the denominator, whether anyone has written down what counts as “held”, and whether the calendar status is set by a human after the fact rather than by the meeting tool. Fix the counting before you spend a dollar on fixing the rate.

Which lever should I change first?

The 1-day SMS that asks for a reply, if you do not already have one. It is a single scheduled send with the best controlled evidence behind it, and the replies tell you hours in advance which calls are at risk. Fix the denominator in the same week, or you will not be able to prove it worked.