A line forms at the counter every day around noon. Two people ordering at once means one cashier, one screen, one conversation at a time, and everyone behind them checking the clock. Somebody suggests a kiosk. Somebody else points out what one costs, and the conversation stops there, usually before anyone's actually worked out whether it would pay for itself.
That's the right question to stall on, because kiosks are one of the few restaurant tech purchases where the cost is upfront and specific and the payoff depends entirely on how long your line actually gets.
What a kiosk is actually replacing
A self ordering kiosk doesn't take orders faster than a good cashier. It takes orders in parallel. One cashier processes one order at a time no matter how skilled they are. Three kiosks process three orders at once, which is the entire value proposition: not speed, throughput. That distinction decides everything else in this post.
It also removes a specific kind of friction: the customer standing at a register while three people behind them read the menu board and haven't decided yet. A kiosk lets the undecided customer step aside without holding up the line, and that alone can shave real minutes off a lunch rush at a busy counter.
The cost side, honestly
Commercial self ordering kiosk hardware commonly runs in the $2,000 to $5,000 range per unit, with ordering software licensed on top of that, and most small operators need at least two units to see the parallel-order benefit at all. That's a real number to plan around, not a rounding error, and it's the reason this decision splits so cleanly by volume: the math that makes three kiosks obvious for a 200-cover-a-day QSR doesn't exist for a 40-cover neighbourhood spot.
Before buying hardware, most small restaurants already have a cheaper version of the same idea sitting in their pocket: a tablet mounted at the counter running the same ordering flow customers use to order ahead. It's not dedicated kiosk hardware, and it doesn't look as polished as a built-for-purpose unit, but it does the same job, taking an order without a cashier typing it in, for the cost of a tablet stand instead of a few thousand dollars a unit. We run this for Eatz operators who want the counter-order benefit before committing to real kiosk hardware: the same QR and counter ordering flow that works for dine-in and walk-up orders sits on a mounted tablet just as well as it does on a phone.
The line-length test
You don't need to buy anything to answer this. Run it for a week:
- At your two busiest lunch or dinner hours, count how many people are standing in line behind the person currently ordering. Do this at the actual peak minute, not an average across the shift.
- Time how long a full order takes at your register, from "hi, what can I get you" to payment complete. Most counter orders run 60 to 90 seconds once someone's decided what they want.
- Multiply your average line length by that per-order time. That's roughly how long your slowest customer is waiting just to start ordering, before their food is even made.
- If that number is regularly under two minutes, a kiosk is solving a problem you don't have. If it's routinely five minutes or more at your actual peak, you have a throughput problem a second order-taking path would fix.
Do this before you price out hardware. The line-length number tells you whether you're shopping for a kiosk or shopping for a tablet stand.
Where it pays off, and where it's dead weight
It earns its cost at high-volume counter-service concepts: quick service, fast casual, anywhere the daily order count is high enough that even a modest per-order time saving multiplies into real minutes of line reduction across a shift. It also helps at spots with a complicated menu full of modifiers and combos, where a screen lets someone build their order at their own pace instead of talking a cashier through it under pressure.
It's dead weight at a full-service restaurant where servers take the order at the table, at a low-volume counter spot where the line rarely gets past two or three people, and anywhere cash is still a big share of transactions, since kiosks add the most value on card and mobile payment, not cash handling. Buying kiosk hardware to solve a line that's five people deep twice a week is spending real money on a problem that shows up for twenty minutes.
The honest boundary
If your line-length math says the wait is real and daily, a purpose-built kiosk is worth pricing out, and it'll pay for itself faster at higher volume. If the math says the wait is occasional or short, start with a mounted tablet running your existing ordering flow before you spend kiosk money on a problem that only shows up during a couple of rushes a week. Plans for running that ordering flow are on the pricing page, and if table service is more your situation than a counter line, table ordering covers the equivalent math for a dine-in room instead of a walk-up counter.
Either way, price out the tablet option first. It's the cheapest way to find out if the kiosk idea earns its keep before you commit to the hardware version of it.
Frequently asked questions
Do self ordering kiosks actually increase revenue?
Mostly through throughput, not persuasion. They let more orders get taken at once during a rush, which means fewer customers walk away from a long line. They don't meaningfully upsell more than a well-trained cashier does.
How much does a restaurant self ordering kiosk cost?
Commercial kiosk hardware commonly runs $2,000 to $5,000 per unit, plus software licensing, and most operators need at least two units to see the parallel-order benefit. A mounted tablet running an existing ordering flow costs a fraction of that but isn't purpose-built kiosk hardware.
Do I need a kiosk if I already have QR ordering?
Not necessarily. QR ordering already lets customers order without a cashier, from their own phone, for dine-in or counter pickup. A kiosk adds a shared screen for customers who don't want to use their phone, which matters more at high volume than at a quiet counter.
What's the fastest way to know if a kiosk would pay off?
Time your line at actual peak minutes, not average traffic. If the wait to start ordering regularly runs past two minutes at your busiest times, there's a throughput problem worth solving. If it doesn't, a kiosk is solving something that isn't costing you much.
Can a kiosk replace my cashier entirely?
For some concepts, mostly. For others, it just adds a second path alongside a person. Cash-heavy counters, older customer bases, and complex custom orders all tend to keep a human option running alongside any kiosk or tablet.