How to reduce no-show appointments
without confirming people to death

Written for sales directors and ops leads whose reps keep pulling up to empty houses: the three-layer confirmation cadence, the script, the measurement, and what to do when the real problem is the sets.

How do you reduce no-show appointments?

Confirm at the point of setting with a calendar entry, a named rep and a repeat-back. Call live the day before (live reminders beat automated ones in the published evidence) and send a short day-of text with logistics. Give wobblers a reschedule route rather than an exit. Then measure sit rate by setter and lead source, because chronic no-show is usually a set-quality problem, not a reminder problem.

Where no-shows come from

The appointment usually dies days before the doorstep.

A no-show is rarely a decision. Almost nobody sits at home at 4:25 watching your rep pull up and chooses not to answer the door. The appointment died earlier, usually somewhere close to the moment it was set, and the doorstep is just where you find the body.

Five causes account for most of what we see on the floors we run and audit:

  • Low commitment at the set. The prospect said yes to end the call, not to meet your rep. Nothing was written down, nothing was agreed beyond a vague slot.
  • Time distance. Commitment decays with every day between set and sit. In the largest dataset we could find, 46,655 ophthalmology appointments at the University of Virginia, the no-show rate in its resident clinic ran at 9.1 percent when the appointment was zero to two weeks out and 38.3 percent at six months out. Different industry, same animal: distance kills intent.
  • Life. Shifts change, kids get sick, cars fail. This is the structural remainder, and no cadence removes it.
  • Buyer remorse. The enthusiasm on the call cools overnight, a spouse asks what it costs, and silence feels easier than a cancellation call.
  • Weak expectation-setting. The prospect does not know who is coming, how long it takes or what to have ready, so it never felt like a real appointment. Things that do not feel real are easy to skip.

The causes matter because each layer of the confirmation architecture below exists to kill a specific one of them. Skip a layer and its cause survives.

Three layers, three different jobs

One text the night before is the weakest version of a good idea.

Most floors run one confirmation touch, and it is usually a text. The evidence says that is the least effective way to do it. Hasvold and Wootton's 2011 systematic review of 29 reminder studies (healthcare, because healthcare measures attendance obsessively) found reminders removed roughly a third of baseline non-attendance, and that live phone calls beat automated messages by a meaningful margin: about a 39 percent relative reduction against about 29 percent. A human on the phone can hear a wobble. A text cannot.

So the confirmation cadence we build for appointment-setting clients has three layers, each aimed at a different cause from the list above.

LayerWhenChannelIts job
At-set confirmationInside the set callLive, same callTurn a polite yes into a commitment
T-1 confirmationThe day beforeLive call, text only as fallbackFind wobblers while the slot can still be refilled
Day-of touchTwo to three hours aheadText, or a 30-second callMake it concrete: rep name, arrival window, what to have ready

At the set

The cheapest confirmation is the one inside the set call, because you will never have more of the prospect's attention than you do right then. The setter gets the appointment into the prospect's own calendar while still on the line ("grab your phone, I'll wait"), names the rep who is coming, and has the prospect say the time back. Repeat-back sounds like theater until you see it measured. In a UK primary care study, asking patients to repeat appointment details back cut missed appointments by 3.5 percent, having them write the appointment card in their own hand cut 18 percent, and the full package of small commitment devices ran at just under 32 percent (Martin, Bassi and Dunbar-Rees, Journal of the Royal Society of Medicine, 2012). None of that costs meaningful talk time.

The day before

A live call, made by a person who owns the diary. Its job is not reminding, it is detection. "Is 4:30 still good for you?" answered a beat too slowly tells you things no delivery receipt ever will, and live is the only channel that converts a brewing no-show into a reschedule on the spot. Send the text as well if you like, but the setups we audit that lean on SMS alone are running the automated arm of the evidence and getting the automated result.

On the day

Logistics, nothing else. The rep's first name, the arrival window, one thing to have ready. A visit that feels arranged gets honored more often than a vague promise does. If the T-1 call never connected, the day-of touch escalates back to a phone call instead of a text.

See whether your Five9 can run this

Our free 17-question Five9 Health Check covers campaign structure, dispositions and list hygiene, the exact plumbing a confirmation cadence depends on. Ten minutes, and you keep the findings either way.

Take the Health Check

An appointment confirmation call script that earns the sit

Assume the appointment stands, hand over specifics, offer an out that lands on a reschedule.

Confirmation scripts fail in two directions. Grovel ("just checking you still want to go ahead") and you invite cancellation. Steamroll ("calling to confirm Mark WILL be with you at 4:30") and the prospect who has gone cold ghosts rather than argues. An appointment confirmation call script has three jobs: assume the appointment stands, make it concrete, and give the wobbler a route that ends in a new slot instead of an exit.

An original example, home improvement flavor, the day before the visit:

"Hi Sarah, it's Jess from Apex. I'm squaring away Mark's diary for tomorrow. He's got you down for 4:30, and he'll want a proper look at the back windows you mentioned. Takes about an hour. Is 4:30 still solid for you?"

If yes: "Perfect. He'll text when he's twenty minutes out. If anything shifts, call me on this number and we'll move it, no drama."

If she hesitates: "Honestly, if tomorrow's got complicated I'd rather move you now than have Mark sat outside an empty house. I've got Thursday at 6 or Saturday at 10. Which suits?"

Notice what is doing the work. "The back windows you mentioned" restates value in the prospect's own words, one clause, not a pitch. A named rep plus a duration makes the visit feel real. The out is framed around not wasting Mark's evening, which most people respond to better than pressure, and it arrives pre-loaded with two specific alternatives so the path of least resistance is a new slot rather than a vague "I'll call you back." The one thing this script never does is ask whether they still want the appointment. That question converts private wobbles into cancellations for free.

Reschedule beats cancel, and cancel beats silence

Every confirmation outcome sits on a ladder, and the script should push people up it.

Train confirmers on a simple ladder of outcomes. A kept appointment is the goal. A reschedule keeps the commitment alive, and in our experience running these desks a decent share of first reschedules go on to sit, though we will be straight with you that we have no published number to hang on that. A cancellation with notice at least returns the slot: the diary refills, the route gets redrawn, the rep sells somewhere else. Silence is the only genuinely worthless outcome, and every script choice should move the prospect at least one rung up from it.

Two rules keep rescheduling honest. First, always offer two concrete slots, never "when suits you," because open questions stall and stalled prospects ghost. Second, cap it. A second reschedule gets flagged, and the third conversation is a different conversation: "It sounds like the timing isn't right just now, shall we leave it with you to come back to us?" That feels counterintuitive to sales managers, but a prospect who has burned three slots is consuming diary you could sell twice over, and some of them, released politely, come back warmer.

Dispositions carry the ladder into your data: Confirmed, Rescheduled with the new date written back, Canceled with a reason, No Contact. If your confirmation outcomes all land in a generic callback bucket, none of the measurement below is possible.

The measurement that settles arguments

One aggregate sit rate is nearly useless; the cuts are where the answers live.

Sit rate is kept appointments divided by appointments set, over a window long enough to be stable. It belongs on the wall next to revenue. The aggregate number tells you that you have a problem; the cuts tell you where it lives.

CutWhat it isolates
By setterSet quality. Two setters on identical leads with a 20-point sit gap is a coaching finding, not a lead finding.
By lead sourceLead quality. Price every source on cost per sit rather than cost per lead and some cheap sources get expensive overnight.
By confirmation completionCadence value. Compare sit rate where the T-1 call connected against where it never did. The gap is what the cadence is worth on your floor, in your own data.
By days from set to appointmentTime distance. If sit rate falls off a cliff past a week out, diary lead time is your no-show generator and the fix is capacity or triage, not more reminders.

Then do the cost math once, roughly, in front of whoever owns the budget. The inputs here are illustrative, so substitute your own. A rep at $60 an hour fully loaded, spending three hours per appointment including travel, costs $180 per doorstep. A floor setting 100 appointments a week with a 30 percent no-show rate is buying 30 dead visits: $5,400 a week, call it $270,000 a year, before counting the revenue those slots displaced. We have yet to walk a home improvement owner through that arithmetic and see the confirmation desk stay unfunded.

Automating the cadence on Five9

Appointment date fields plus dispositions, so the list builds itself.

None of this survives contact with a busy floor if it depends on someone remembering to export a list every afternoon. The setups we audit usually have exactly that: a spreadsheet, one person who knows how it works, and a cadence that dies whenever they are on leave. The whole architecture runs natively on the dialer.

The mechanics on Five9:

  1. Write appointment date, time and confirmation status onto the contact record as fields at the point of set, from the setter's disposition or your CRM sync.
  2. Run a dedicated confirmation campaign against a list filter: appointment date equals tomorrow, status not yet confirmed. Five9 outbound campaigns support predictive, power, progressive and preview modes; for a confirmation desk we set power or preview at a gentle pace, because these are warm, known contacts, not cold list.
  3. Let dispositions do the routing. Confirmed is terminal. Reschedule writes the new date back so the record re-enters the same filter on the right day. No Answer carries a disposition redial timer so the record retries that evening, which is precisely what Five9's redial timers exist for.
  4. Hang the day-of text off the same fields through Five9's SMS and Workflow Automation tooling, which Five9 positions for appointment reminders among other proactive notifications.
  5. Disposition everything. The sit-rate-by-confirmation-completion report is a join between confirmation outcomes and appointment outcomes, and it only exists if both sides are captured.

One housekeeping note: automated texts carry consent obligations on both sides of the Atlantic, and the rules have moved in the last couple of years, so check what your lead forms actually captured before switching the SMS leg on. We cover the detail in our lead generation consent guide.

If your campaigns, filters and dispositions cannot carry this today, that is a configuration gap rather than a platform limit, and closing gaps like it is the bread and butter of our Five9 optimization work.

The floor under the number

Some no-show is structural, and chasing zero over-invests.

The honest answer on how low no-show goes: not zero, and not especially close. The reminder literature's ceiling is instructive. Across those 29 studies, reminders removed about a third of baseline non-attendance with everything working. Commitment devices at the set stack on top, but illness, emergencies and genuinely changed circumstances remain, and no cadence touches them. That is the structural remainder we flagged at the start, and it is real.

Past a certain point, extra confirmation spend inverts. A fourth touch does not rescue committed prospects, it irritates them, and the desk hours would return more if spent refilling canceled slots or coaching set quality. We have watched a floor chase its last three points of no-show with an escalating reminder sequence while its weakest setter ran a sit rate 20 points under the floor average, completely untouched. The marginal hour had an obvious better home.

So set a working floor from your own confirmed-appointment sit rate by segment, hold the cadence at three touches, and once you are near that floor, move the effort upstream.

When it is not a confirmation problem at all

A stubborn no-show rate with a clean cadence behind it points at the sets or the leads.

The tells are in the cuts. No-shows concentrated in one lead source mean that source's real intent is not what the invoice says: reprice it on cost per sit or drop it. No-shows concentrated under one or two setters mean you should go and listen to their set calls, where you will usually hear appointments booked at people rather than with them, vague slots agreed to end the call, quota pressure doing the talking. In home services, where the no show rate on canvassed and aggregator leads can dwarf anything a reminder fixes, this cut matters more than all the others combined.

Home improvement floors add a specific failure mode: the one-legger, an appointment set without both decision-makers present, which either sits and cannot close or no-shows when the absent partner vetoes it overnight. Requiring both names at the set costs some volume and repays it in sit rate and close rate. We cover set quality for that industry properly in our home improvement appointment setting guide.

Speed matters upstream too. Leads worked while intent is fresh set appointments with shorter lead times and stronger commitment, and given what time distance does to attendance, speed to lead is a no-show intervention wearing a different name. Pay setters on sits rather than raw sets and most set-quality problems start fixing their own incentives.

Asked & Answered

What is a normal no-show rate for home services appointments?

Nobody publishes reliable cross-industry numbers, so treat any confident benchmark with suspicion. On the home services and field sales floors we run and audit, the spread between the best and worst setter on the same floor is usually wider than the spread between companies, which tells you most of the variance is set quality rather than industry. Build your own baseline instead: four weeks of sit rate cut by setter and by lead source, then improve against that.

Should appointment confirmations be a phone call or a text?

Both, doing different jobs. The strongest evidence comes from healthcare, where a 2011 systematic review of 29 studies found live phone reminders cut non-attendance by roughly 39 percent of the baseline rate against about 29 percent for automated messages. The day-before touch should be a live call because a human can hear hesitation and convert it into a reschedule on the spot. Texts carry the day-of logistics: rep name, arrival window, what to have ready.

When is the best time to confirm an appointment?

Three moments: at the point of setting, the day before, and on the day. The at-set confirmation does the most work because attention is at its peak, and small commitment devices like a repeat-back are measurably effective. The day-before call catches wobblers while the slot can still be refilled. The day-of text is pure logistics. For a single reminder, the research found little difference between a day ahead and a week ahead, which is why layering beats timing.

Does asking someone to confirm just invite them to cancel?

The cancellation is already in them by that point. The question is whether you find out the day before, when the slot can be refilled, or on the doorstep, when the cost is already sunk. What genuinely invites cancellation is a groveling script. Assume the appointment stands, ask whether the time is still solid, and if you hear a wobble, offer two specific alternative slots so the easy exit is a reschedule rather than a no.

How do I calculate what no-shows cost?

Fully loaded rep cost per hour, times hours per appointment including travel, times missed appointments per week. As an illustration with round numbers: at $60 an hour, three-hour visits and 30 no-shows a week, that is $5,400 a week in dead cost before counting the revenue those slots could have produced. Swap in your own figures. The shape of the result is usually enough to fund a confirmation desk on its own.

How do I automate appointment confirmations in Five9?

Store appointment date, time and confirmation status as contact fields, then run a dedicated confirmation campaign against a list filter for tomorrow's unconfirmed appointments. Dispositions handle the routing: confirmed records leave the campaign, reschedules write a new date back and re-enter the filter on the right day, and no-answers retry on a disposition redial timer. The day-of SMS hangs off the same fields through Five9's messaging and workflow tools. Check your consent capture before enabling automated texts.

Why are my no-shows concentrated in one lead source?

Because that source's real intent is lower than its price implies. Sit rate by lead source is the cut that exposes it: a source can produce cheap leads and expensive sits at the same time. Reprice every source on cost per kept appointment rather than cost per lead, feed the numbers back to the vendor, and drop sources that stay upside down. No confirmation cadence rescues appointments that were never really wanted.

Is a reschedule really better than a cancellation?

Yes, up to a point. A reschedule keeps the commitment alive, and in our experience a worthwhile share of first reschedules go on to sit. A cancellation with notice at least frees the slot for refilling, which still beats silence, the only truly worthless outcome. Cap it though: flag second reschedules, and treat a third as a signal the timing is wrong. Politely releasing a serial rescheduler protects your diary, and some come back warmer later.

References

Keep Reading

Bring us the number that will not move

We run outbound floors and confirmation desks daily, and we can usually tell within a week whether a no-show problem is cadence, sets or leads. You get a dedicated pod that knows your diary, not a ticket queue.

Talk to us