There is a specific kind of bad week that shows up in every stylist forum eventually: someone adds up the no-show fees their software was supposed to have collected over a year and finds that a large share of them never actually charged. The policy was set. The card was on file. The client didn’t show. And the money isn’t there.

This post is about why that happens, because the mechanics are genuinely not obvious, and nobody selling you booking software has much incentive to explain them. Nothing here is specific to any one platform — it’s how card payments work, and it’s the same on all of them.

Three different things that all sound like “we have their card”

Saving a card means storing a token that lets you attempt a charge later. It proves the card existed and was valid at the moment it was saved. It proves nothing about whether money will be there on Tuesday. A saved card is permission to try, not a promise of funds.

Authorizing a card — a preauthorization, or “preauth” — is a real-time request to the bank that sets aside a specific amount for a specific window. The bank checks the funds, holds them, and hands back a code. The money is not moved; it’s reserved. Later you capture that authorization and the reserved money actually transfers. Authorizations expire — typically in about a week, sometimes less on debit cards — and if you don’t capture in time, the hold falls off and you’re back to attempting a fresh charge.

Charging a card moves money now.

Almost every no-show policy in this industry is built on the first one. You save a card at booking, and when the appointment is missed, you attempt a charge — often days later. Between those two moments, everything that can go wrong with a card can go wrong.

Why the attempt fails

In rough order of how often it happens:

Insufficient funds. The most common one, and the most predictable: the client who no-shows a $60 appointment is disproportionately the client whose account is at $12.
The card is gone. Reissued after a fraud alert, expired, cancelled, replaced when the bank shipped a new chip. Card-on-file updater services fix some of this automatically, and only some.
The bank declines an unexpected merchant. A charge from a business the cardholder hasn’t visited in a while, for an amount they didn’t approve today, is exactly what fraud models are tuned to reject.
The client disputes it. This one is worth sitting with: a saved card with no explicit, recorded, at-booking agreement to a specific fee is weak evidence in a chargeback. The client says they never agreed; you say the policy was on the page. Whoever documented it better tends to win.
Nobody actually ran it. Plenty of “automatic” no-show fees require a human to mark the appointment as a no-show first. If Saturday gets away from you, the fee never fires.

Notice that a preauthorization at booking would have caught the first three at the moment of booking — while the client is still in front of you, still able to try another card, and still willing to.

So why doesn’t everyone preauthorize?

Because it has real costs, and anyone who tells you otherwise is selling something.

It’s visible to the client. A $50 hold on a debit card reduces the balance they can spend, immediately, for a service next week. For some clients that’s fine. For others it’s the reason they book somewhere else, and you will not get an email explaining why.
It expires. Book three weeks out and the authorization is dead before the appointment. You’d have to re-authorize closer to the date, which means another hold and another chance to fail.
It’s more machinery. Holds, captures, releases, and expiries are more moving parts than a deposit, and every moving part is a thing that can confuse a client at the worst moment.

Preauthorization is a real tool, and it is not the free answer it sounds like.

What actually collects

Ranked by what reliably ends with money in the account:

A deposit taken at booking. It’s not a hold and not an attempt — it’s a completed payment, made while the client is motivated. It also does the thing a fee can’t: it changes the odds they show up at all. A booking someone paid for is a booking they remember. This is the single highest-leverage change most shops can make, and it’s why we wrote a whole post on introducing deposits without scaring off your regulars.
A deposit plus a clearly agreed, documented policy. The deposit collects, and the record of what the client agreed to — shown before they confirmed, timestamped — is what makes the rest of the fee defensible if you pursue it.
A preauthorization at booking, for short booking windows and high-value services, where the hold is worth the friction.
A saved card and a charge attempt afterward. This is the most common setup in the industry and the weakest of the four. It works often enough to feel like it works, and fails silently enough that you don’t find out until you total it up.

The uncomfortable arithmetic

If you’re going to take one thing from this: run the numbers on what your no-show policy actually collected, not what it was supposed to. Pull the last twelve months. Count the appointments marked no-show. Count the fees that settled. The gap between those two numbers is the real policy — the one your business is actually operating under.

Most owners have never run that comparison, and the number is usually worse than they’d guess. If you want the other half of the math — what those missed appointments cost you before any fee enters the picture — our no-show calculator and the methodology behind it are both public, assumptions and all.

Where Parlor stands, stated plainly

Parlor takes deposits at booking — a percentage of the ticket or a flat amount, set per service — and charges no-show fees to the card on file, with the policy shown to the client before they confirm. That’s options 1 and 2 above, which is where we think the leverage actually is.

Parlor does not do preauthorization holds today. It’s on the list, and this post exists because the mechanics are worth understanding whether or not you ever run a hold — but if someone tells you we do, they’re wrong, and we’d rather you learn that here than from a support reply.