Why Does Every Checkout Screen Ask for a Tip Now?

Why does a screen ask you to tip on a coffee you carried to your own table? Look at what it suggested. One dollar. Nobody behind that counter chose that number. It came with the software. That isn't a complaint about the shop. It's a description of how the number got there. Take Square, whose readers sit on a lot of American counters. In the Square app, tipping is a switch on a setup page. And when the shop flips it on, the amounts are already written. Square publishes them. There's also a rule underneath them that almost nobody in the shop has ever read. A line at a certain dollar amount where the screen quietly changes the question it's asking you. In 2009 in New York, two rival companies bolted their screens into taxi cabs, made slightly different choices about what to print on their own buttons, and accidentally ran an experiment on millions of people. Nobody meant to run it. It ran anyway. Start with the rule. Above a certain size, the buttons are percentages. But when the bill drops below ten dollars, the percentages disappear and the screen asks for whole dollars. Instead, the middle button is two. On a four dollar coffee, that's half the drink. Half a coffee for handing you a coffee. Nobody wrote that for your cafe. It's the preset. And it's the same preset waiting at the bakery, the bike shop, and the place that cut your keys. So here's the obvious explanation. And it's the one most of us carry around. Businesses did this to us. Labor got expensive. Prices were already uncomfortable. And rather than raise wages, somebody figured out how to hand the wage bill to the customer one tap at a time. It's a tidy story. It even describes some restaurants accurately. But follow the dollar, and it starts coming apart. Tap the button, and that dollar never gives the shops money. Under federal law, tips are the property of the employee who received them. The owner can't take a cut. Neither can a manager. Now, there's a real caveat here, and it's the strongest thing the wage story has going for it. In most states, an employer is allowed to count part of an employee's tips toward the minimum wage it owes them. So tips genuinely can lighten a payroll. Michael Lynn at Cornell, who has spent decades studying this, has an answer to that. Restaurants compete hard enough that owners mostly hand the saving straight back to customers as lower menu prices rather than pocketing it. Either way, it's an argument about places with tip staff. It isn't an argument about the card reader at a dry cleaner. And meanwhile, the card companies take their percentage off the whole total, tip included. That's in the processor's own published fee terms. Bigger tip, bigger fee. That part isn't strategy. It's arithmetic. There's stranger problem the wage story, which is where the prompts turned up. Furniture showrooms, dry cleaners, drive-thrus, self-checkout, where the machine is asking you to tip the person who scanned your groceries, and that person is you. And the thing people assume happened, that we all caved, that tipping at counters quietly became normal, mostly didn't. Pew Research Center put this to Americans in 2023. Among people who eat at sit-down restaurants, 92% said they always or often tip. Among people who buy coffee, it was about 1 in 4. Nobody's habits migrated to the counter. The counter started asking, which leaves the real question. If it isn't a change in custom, and it isn't a wage strategy, why did it appear everywhere in the same handful of years? Because thousands of unrelated businesses bought the same handful of checkout systems, and the prompt was already in the box. Pew points at the spread of point-of-sale tablets, apps, and kiosks. The Federal Reserve Bank of Richmond, writing about tipping in 2024, points at the same thing. Small businesses buying more capable card terminals that happened to arrive with a tipping option attached. It spread the way a software update spreads, because that's what it was. Which brings us back to the cabs. Every yellow cab in New York had a screen in the back seat, and the screens came from competing companies. Two economists, Karim Haggag and Giovanni Pace, found the experiment sitting inside the city's own trip records, and they found it twice. One of those companies had its own line. Once a fare grew past a certain size, its screen switched to a higher set of suggestions, so the pair simply compared wide rides that landed just under that line against rides that landed just over. Then they checked it a second way. One company's amount sat higher than the others, and riders don't get to choose their cab. At LaGuardia, you queue, and you take whichever one reaches the front. Both times, the riders looking at the higher suggestions left more. Somewhere between about 20 and 30 cents extra a ride at the airport. Comparison at the low end of that. The fare line won at the high end, which sounds like nothing and is nothing until you're the driver. The authors worked out that across a typical driver's year behind the wheel, it came to between roughly $250 and $375. Nobody's mind was changed. A button moved. There was a second finding, and it's the interesting one. Faced with the pushier suggestions, more riders left nothing at all. Not less. Nothing. Push a default too far, and some people stop negotiating and start refusing. A default only has that kind of grip when the decision is fat, small, and slightly uncomfortable. That's the other half of why the screen works, and it's the half nobody designed on purpose. Tipping used to happen unobserved. You left coins on a table. Nobody was watching. You wrote a number on a slip and walked out before anyone read it. The whole ritual was arranged so you never had to defend the figure. Then the screen arrived, and it doesn't face the room. It faces you, held by a person standing three feet away who will see what you tap, with someone behind you in line waiting for you to finish. Researchers who study digital tipping describe this bluntly. The point-of-sale screen broke the privacy that tipping used to have. What that does to the number, honestly, is contested. A 2025 study by Nathan Warren and Sara Hansen found that losing privacy is that privacy pulls in two directions and goes in one direction. Feeling watched makes people feel less in control, which pushes the tip up, and less generous, which pushes it down, so the effect on the amount itself is murky. What they did find clearly was that customers who felt watched were less likely to come back. Whatever the screen is buying, it isn't goodwill. So, back to the counter. Coffee in your hand, card in the reader, the screen swinging around to face you. Here's what's actually happening in that moment. You are not being asked what the coffee was worth to you. You're being asked in public to answer a question that a payment company wrote in a settings menu in front of somebody who has watched a few hundred people answer it since breakfast. The barista didn't write it. The owner switched it on and probably never touched the numbers. And the money really does reach the worker. That part is real, and it's the only part of this that anybody planned. Here's my opinion, and it's an opinion. We've spent a few years arguing about generosity when the thing that actually changed was a settings menu. And I don't think a preset has earned the authority of a custom. A custom is something people build between themselves, badly and slowly, over decades. This one was written once, switched on, and left alone by everyone who came near it. Although, and this is what I keep getting stuck on, presets harden. The percentage of people who rolled their eyes 40 years ago is the polite floor now. Leave a number on a screen in front of enough people for enough years, and it stops being a suggestion and turns into the thing everyone assumes everyone else does. Which means the strange part isn't the screen asking. It's that a number nobody in the room chose is slowly becoming a number the room expects. Thanks.