A deposit is a trade, and most people never price it

The trade nobody prices

There's a section in our no-show playbook that ends badly. It says deposits probably work, notes that the published evidence is thinner than the confidence around it, and moves on. Which is not much use if you're the one deciding whether to put a card form in front of your booking page on Monday morning.

So here's the part I left out. A deposit is not a free anti-no-show device that you either switch on or don't. It's a trade. You get fewer people failing to turn up, and you pay for it in people who start booking and stop when the payment step appears. Both effects are real. Both are measurable. Almost every deposit discussion I've read online prices the first one and ignores the second entirely.

The question isn't "do deposits reduce no-shows." They do. The question is whether the no-shows you prevent are worth more than the bookings you lose, and that depends on numbers specific to your business.

The arithmetic, with your numbers in it

Four numbers decide it.

• v — what a completed appointment is worth to you
• s — your current no-show rate
• s′ — your no-show rate once a deposit is in place
• a — the share of would-be customers who abandon at the payment step
• d — the deposit amount

Without a deposit, every hundred people who start a booking are worth 100 × (1 − s) × v. With one, they're worth 100 × (1 − a) × [(1 − s′) × v + s′ × d] — fewer people book, but the ones who no-show leave the deposit behind.

Set those equal and you get the abandonment rate at which the deposit stops paying for itself. Three worked examples, with assumptions I'm making up for illustration rather than measuring — plug in your own:

Read the middle row again, because it's the one that surprises people. If your no-show rate is already low, a deposit has almost no room to earn its keep — lose three bookings in a hundred to the payment step and you're behind. Deposits are a high-no-show instrument. Businesses with a disciplined client base introduce them, watch the calendar thin out slightly, and never connect the two events.

Two corrections to the model, both in the same direction:

If you can refill a cancelled or missed slot, a no-show doesn't cost you the full v — it costs you the gap. That lowers the value of the deposit and tightens the break-even further. And if the deposit is applied to the final bill rather than kept, you haven't gained d on a no-show unless your terms say you keep it, which is a separate question I'll get to.

So the honest version of the model is more pessimistic than the table. I'd still rather publish the pessimistic version than the one that makes the deposit look like a free win.

A deposit and a no-show fee are not the same product

These get used interchangeably and they behave nothing alike.

A deposit is a prepayment. The customer pays part of the price now, it comes off the bill later, and the money has already moved before anything goes wrong. Customers understand it because they've paid deposits for hotels and dentists. The friction is entirely at booking time.

A no-show fee is a penalty. You store a card at booking, charge nothing, and charge it after they fail to appear. Booking is frictionless. Everything else is harder: you need the card saved and authorised for later use, you need consent that holds up when it's disputed, and you need the nerve to actually charge it — which, in practice, is where most no-show fee policies quietly die. A fee that everybody knows you never charge has the deterrent value of a sign.

Which should you use? If your problem is people who book and vanish, a deposit works better, because it moves the money while they still want the appointment. If your problem is losing bookings and you can't afford friction at the top of the funnel, a stored card with a genuinely enforced fee is the tool — and you have to enforce it.

What I wouldn't do is both, at full strength, for a first-time customer. That's two separate asks for money before you've done any work.

Where the booking actually dies

Abandonment isn't evenly spread across the flow. It clusters at one step, and it's rarely the price itself.

The things I'd look at first, roughly in order of how often they turn out to matter: the deposit amount appearing for the first time on the payment screen, after the customer has already picked a time and entered their details. A card form on a phone, on a page that doesn't look like the salon's website. A processor descriptor the customer doesn't recognise. And a required field they didn't expect — address lines are a classic, because nobody expects a haircut to need a billing address.

The fix for most of that is sequencing, not discounting. Say the deposit amount on the service, next to the price, before anyone picks a slot. People who won't pay it drop out before investing five minutes, which costs you nothing and keeps your abandonment metric honest. People who will pay it arrive at the payment step having already agreed to it.

I'd rather lose a booking at step one than at step four. Same lost booking, much better information.

The customers I'd exempt

Here's an opinion you can disagree with: charge new customers, not returning ones.

Your no-show risk is not uniform. A client who has come in six times is not the person costing you Tuesday afternoons, and asking them for money up front is a small insult with a real cost — they notice, and some of them shop around after noticing. The deposit is a screening tool for people you don't know yet, and screening someone you already know is just friction.

Practically: put the deposit on first bookings, on long or high-value services, and on the slots that hurt most when they go empty. Leave repeat clients booking a routine appointment alone.

What would change my mind: if your no-shows are concentrated among regulars rather than newcomers, ignore all of this and do the opposite. Check before you assume. A rough count of the last thirty no-shows, sorted into "first visit" and "not", is twenty minutes of work and it settles the question for your business rather than in general.

What happens when someone disputes the charge

A percentage of deposits and almost all no-show fees will be disputed eventually. Worth knowing how that goes before it does.

The customer contacts their bank, the bank reverses the payment, and you're asked to provide evidence. For a deposit on a service that was never delivered, "they didn't turn up" is a weaker position than it feels like, because from the card network's view you're charging for something that didn't happen. What tends to help: the policy text shown at the moment of booking rather than buried in a page nobody opened, a timestamped record of the customer accepting it, and the booking confirmation with the terms in it. What doesn't help: a policy on your website that the customer never had to pass through.

Two things to be realistic about. You will lose some disputes you should win — that's the cost of doing business with cards, not a sign you did it wrong. And this is one of the places where I'd rather point you at a lawyer than pretend a blog post is enough: consumer protection rules on prepayments and cancellation charges vary by country, and some of them limit what you can keep regardless of what your terms say.

Run it on half your services first

You don't need a data team to answer this. You need a split.

The plumbing, and the part we don't control

Deposits are part of the booking flow rather than a plugin: you set a deposit on a service, the customer pays it at checkout through Stripe or Shopify Payments, and the balance sits on the order until the appointment is done. Cancellation terms live next to it, which matters for the dispute evidence above — the customer passes through the policy on the way to paying, rather than being expected to find it. The details are on deposits and cancellations and in the orders and payments feature page; if you're on Shopify, the platform-specific version of this is taking deposits on Shopify bookings.

The limit worth stating plainly: we're not a payments company. Deposits run through Stripe or Shopify Payments, which means the dispute process belongs to them, and if a chargeback goes against you, there is nothing we can do about it from our side. Anyone who tells you their booking software protects you from chargebacks is describing something they don't control either.

What I can't tell you

Your abandonment rate. Not approximately, not as a rule of thumb.

I went looking for a published figure for how many service bookings are lost when a deposit is introduced, and what I found was retail cart-abandonment statistics being reused as if a haircut and a shopping basket behave the same way. They don't — someone who has picked a specific stylist at a specific hour has a level of intent that a shopping cart never has. So the one number the entire decision turns on is the one number I can't hand you, and neither can anyone else quoting a percentage at you with confidence.

What you can do is measure it in your own data in about eight weeks, which is what the split above is for. Start there, with the deposit on your highest-value service where the break-even has the most room, and make the second decision with numbers instead of vibes.