Appointment Deposits and No-Show Fees: Who You May Charge

A patient books a 90-minute crown prep, does not arrive, and does not call. The chair sits empty, the assistant has nothing to do, and vendor estimates put the lost production somewhere between $500 and $1,500. Do that twice a month and it is a real number by year end.

The obvious response is to make the appointment cost something before it happens. Deposits and no-show fees both do that, they are not the same instrument, and one of them is illegal to charge a large group of your patients.

That last point is where most advice on this topic is simply wrong. A widely repeated line says a practice can charge no-show fees in all 50 states as long as the policy is disclosed. For Medicaid beneficiaries that is not correct, and the consequences of getting it wrong are worse than the no-show ever was.

What is the difference between a deposit and a no-show fee?

One is collected before, the other is charged after. A deposit is money taken at booking and credited against the visit, so a patient who attends loses nothing. A no-show fee is a charge raised after a patient fails to attend, which means collecting money from somebody already unhappy.

That difference drives everything else. A deposit is a commitment device: the patient has skin in the game before the date arrives, and attending costs them nothing extra. A no-show fee is a penalty, applied retrospectively, to somebody who may simply have forgotten and who now has a reason to argue with your front desk.

One is paid before anybody is annoyed. The other is invoiced after everybody is.

There is a third instrument practices forget they already have, which is the short-notice cancellation charge. It sits between the 2, applying when somebody cancels inside a window rather than failing to appear at all. It is worth defining because without it patients learn that cancelling 20 minutes before is treated the same as attending, and the schedule becomes just as hard to fill.

Collection difficulty follows the same split. A deposit is already paid. A fee has to be invoiced, chased, and often written off, which means the true yield on no-show fees is far below the amount billed. Practices that have run both usually report the deposit changing behaviour and the fee mostly generating friction.

How much does a broken appointment actually cost?

More than the missed procedure, because the slot cannot be resold at short notice. Dental benchmarks put average no-show rates between 11% and 15%, and healthcare more broadly loses roughly 10 to 30% of scheduled appointments, with many practices around 15%.

The cost is not evenly distributed, which is the useful insight. A missed 30-minute check costs you 30 minutes. A missed implant consult or sedation appointment costs a long block that was staffed, prepared, and impossible to fill on 2 hours notice. Vendor estimates put those at $500 to $1,500 each, and while that figure is modelled rather than measured, your own version is easy to calculate from your fee schedule.

The first 2 are benchmarks. The last 2 are vendor estimates; model your own instead.

That concentration is why blanket policies underperform. Applying a deposit to every appointment annoys hundreds of reliable patients to protect against a handful of expensive misses. Applying it only to appointments over a length or value threshold targets the actual loss, and is far easier to explain to a patient who asks why.

Worth calculating your own number rather than borrowing anyone's. Take the 3 appointment types with the longest chair time, multiply each by the production you would have billed, and count how many of each were missed last quarter. That figure is specific to your practice, defensible in a staff meeting, and usually enough on its own to settle whether a deposit policy is worth the friction.

Before reaching for money as the lever, it is worth confirming the misses are not caused by something cheaper to fix. Practices with weak reminders often see no-show rates fall substantially once reminders and confirmations are working properly, which costs nothing and irritates nobody.

Which patients can you legally charge?

Not Medicaid beneficiaries. Federal rules prohibit billing Medicaid patients for missed appointments, and CMS reads the regulation requiring providers to accept the state payment as payment in full as making a no-show charge impermissible. Missed appointments are treated as a cost of doing business.

This is the part that gets practices in trouble, because the rule is counterintuitive and the marketing around deposits ignores it entirely. Reported penalties run to over $25,000 per incident along with potential exclusion from federal programs, which makes a $50 fee an extraordinarily bad trade.

Medicare is the mirror image and surprises people in the other direction. Charging Medicare beneficiaries for missed appointments is generally permitted, on the condition that you do not discriminate: the same policy has to apply to your non-Medicare patients too. A policy that exempts everyone except Medicare patients is the problem, not the charging itself.

One more constraint worth knowing: the fee cannot be billed to insurance. It is the patient's own responsibility, which is another reason the collection rate on fees disappoints.

What makes a policy defensible?

Disclosure before booking, and consistent application afterwards. A policy the patient agreed to in writing before the appointment existed is enforceable in a way that a sign at the front desk is not. Applying it to some patients and not others undermines it entirely.

Consistency is the requirement people underestimate, and it has teeth in 2 directions. Legally, a policy waived for whoever complains loudest is hard to defend as a policy at all. Practically, staff need a written rule about waivers, or every charge becomes a negotiation at the desk.

Disclosure before booking and consistency afterwards are what make a policy hold up.

It also helps to decide in advance who is allowed to waive a charge. If any staff member can, the policy will erode within a month, because waiving is the path of least resistance when somebody is upset at the desk. If only the practice manager can, the rule survives and the front desk has something to point at that is not their own judgement.

The disclosure itself should be boring and specific: what triggers the charge, how much it is, how much notice avoids it, and how a deposit is applied if the patient attends. Patients rarely object to a clear rule stated in advance. They object to a surprise.

What should be automated, and what must not be?

Automate the disclosure, the deposit collection and the reminder. Never automate the charge itself. Taking money from a patient who did not attend deserves a human deciding whether this particular absence was a genuine emergency, because plenty of them are.

The automated half is straightforward and genuinely useful. A booking flow that presents the policy and takes a card, a stored payment method, a reminder sequence that repeats the cancellation window, and a same-day alert when somebody does not arrive. All of that is administrative and none of it is a judgement call.

The payer question comes first. It is the one that carries real regulatory risk.

The manual half is a single decision: charge, waive, or call. Automating that decision produces the worst possible outcome, which is a card charged automatically after a patient missed an appointment because they were in an emergency room. That is a review, a complaint and a lost patient, all for $50.

The reminder content matters as much as the reminder itself. A confirmation that repeats the cancellation window and the deposit terms turns the policy into something the patient was reminded of twice, which is both fairer and considerably easier to enforce. A bare reminder of the date and time does none of that work.

An AI receptionist fits the automated half well. It can state the deposit policy, take a booking, and repeat the cancellation window on a confirmation call, because none of that requires judgement. It should not be the thing deciding whether a specific absence was excusable, which is the same boundary described in what an AI receptionist does and where it stops.

Where should you start?

With the payer question and 1 appointment type. Confirm which of your payers prohibit missed-appointment charges, then apply a deposit to your single longest or highest-value appointment type and nothing else. That is a small, reversible experiment with a measurable result.

Tell the team why before you tell the patients. Front-desk staff are the ones who will absorb every complaint about a new charge, and a policy introduced without explaining the reasoning gets quietly undermined at the desk within weeks. Ten minutes explaining which appointments are affected, why those ones, and who can waive a charge buys more compliance than any written procedure.

Run it for a quarter and compare the no-show rate on that appointment type against the rest of your schedule. Vendors claim deposits push no-show rates under 3% on high-value appointments, which is a vendor claim rather than a finding, but your own before-and-after on 1 appointment type is real evidence and costs nothing to collect.

Watch the booking rate alongside the no-show rate while you run it, because a deposit can suppress both. If the no-show rate falls but the number of booked long appointments falls further, the deposit is filtering out patients who would have attended, and you are worse off. That pairing is the only honest way to read the experiment.

If the result holds, widen the threshold slowly. If it does not, you have learned that your no-shows are a reminder problem rather than a commitment problem, which is cheaper to fix anyway. For a wider view of where bookings are leaking before anybody misses an appointment, our free Growth Leak Audit covers the whole path.