The food truck problem
Food trucks are a fixture of American cities. They cluster at breweries, office parks, festivals, and weekend markets. The appeal is obvious: gourmet food, quick service, variety for groups with different tastes.
But food trucks create a splitting problem that restaurants don’t. There’s often no receipt. Some are cash-only. Orders happen fast, under pressure, with a line forming behind you. And when a group hits multiple trucks in one outing, the mental math gets much harder.
We couldn’t find a reliable count of cash-only trucks. The Federal Reserve’s payment diary does show that cash is still ordinary for small purchases.
The speed that makes food trucks convenient is exactly what makes splitting difficult. There’s no time to negotiate. No table to sit at and sort it out. Just a window, a line, and pressure to move.
Source: Bayeh, Cubides & O’Brien, “2024 Findings from the Diary of Consumer Payment Choice,” Federal Reserve Financial Services (2024).
Why cash makes splitting harder
In 1998, behavioral economists Drazen Prelec and George Loewenstein published a landmark paper on what they called the “pain of paying.” They argued that payment methods differ in how tightly they tie a purchase to its payment. Credit cards tend to weaken that link. Cash produces tight coupling.
When you hand over physical currency, the purchase and the payment sit close together in your mind. That may be one reason cash splitting can feel more fraught than digital splitting.
Abstract numbers. Payment feels more distant from the purchase.
Physical exchange. The payment stays tied to the purchase.
Apply that to a food truck and you get a specific dynamic. The person who pays cash for the group feels the payment. Everyone else pays nothing at the window, so the IOU that follows can feel less urgent.
Richard Thaler’s research on mental accounting describes how people open and close mental accounts around a purchase. Our reading of the food truck case: once everyone has eaten, the outing feels closed, and the unpaid debts sit in a separate category that’s easier to forget.
Sources: Prelec & Loewenstein, “The Red and the Black,” Marketing Science (1998); Thaler, “Mental Accounting Matters,” Journal of Behavioral Decision Making (1999).
The multi-truck outing
Here’s an example outing at a food truck park: five friends, three trucks, zero shared receipts.
Now calculate who owes whom. Go ahead. Take your time.
George Miller’s 1956 paper put the span of immediate memory at about 7 items, give or take 2. By our rough count, a three-truck outing with five people generates at least 15 variables: who ordered what at each truck, who paid at each truck, which items were shared, who tipped and how much. That’s about double Miller’s figure.
A rough count: 3 trucks x 5 people = 15 potential order combinations. Add tips, tax, and “I’ll get you back” promises, and you could be tracking 20 or more variables. Miller’s figure is about 7. Your brain is overloaded before dessert.
John Sweller’s work on cognitive load points the same way. He found that conventional problem solving uses a large amount of cognitive processing capacity, which leaves less for anything else. He studied learning, not dinner. In a group, our guess at the likely shortcut is “let’s just call it even.”
Sources: Miller, “The Magical Number Seven,” Psychological Review (1956); Sweller, “Cognitive Load During Problem Solving,” Cognitive Science (1988).
The tipping confusion
How much do you tip at a food truck? Many people aren’t sure. In Pew Research Center’s 2023 survey, only 33% of U.S. adults said it’s extremely or very easy to know how much to tip for different services, and counter-service tipping norms are genuinely ambiguous.
The setting changes the answer. Among adults who eat at sit-down restaurants, 92% told Pew they always or often tip. At fast casual restaurants, that falls to 12%. Pew didn’t ask about food trucks. Food trucks occupy an awkward middle ground: more service than a vending machine, less than a sit-down restaurant.
Food truck tipping guide (our suggestions, not survey data)
The digital tip screen has changed this calculus. Payment screens can show preset tips such as 18%, 20%, and 25%. That can add social pressure a cash tip jar doesn’t. But when one person pays for the group, who decides the tip amount?
The person swiping the card often tips based on their own comfort level, not a group consensus. If they tip 25% on a $50 order, that’s $12.50 in tips that gets split unevenly or forgotten entirely when calculating who owes what.
Sources: Pew Research Center, “Tipping Culture in America: Public Sees a Changed Landscape” (2023); Square, “The Restaurant Operator’s Guide to Fees in 2025.”
How groups actually handle it
Without a system, groups default to one of four approaches at food trucks. Each has predictable failure modes.
“Everyone pay your own”
Each person orders and pays at the window. Works only if everyone has cash. Breaks down with any shared items or cash-only trucks when someone doesn’t have bills.
“One person covers it”
Fastest at the window. One card or cash stack. But now one person is the bank, and collection becomes a separate chore.
“Rotate who pays”
Alex gets truck 1, Jordan gets truck 2. Fair in theory if prices balance. Say the tacos come to $14 and the brisket plate to $24: the rotation is already uneven.
“Track and settle”
One person pays at each truck. Log all orders. Calculate exact shares including tip. Settle immediately via app.
Uri Gneezy and his coauthors tested this in a restaurant. The average meal cost 37.3 Israeli shekels when diners paid individually and 50.9 when they split evenly, more than a third higher. “One person covers it” may create a similar dynamic at food trucks: without clear tracking, people can order more knowing the cost will be spread across the group.
Source: Gneezy, Haruvy & Yafe, “The Inefficiency of Splitting the Bill,” The Economic Journal (2004).
The time pressure factor
Food truck transactions happen fast. Trucks run on volume, and the line has to move.
Mental math needs working memory, and the window leaves little to spare. Mark Ashcraft and Elizabeth Kirk found that people with high math anxiety have smaller working memory spans. That led to a pronounced increase in reaction time and errors when they did mental addition while holding a memory load. Their study was not about food trucks or the clock. A line, an order and a group still compete for the same attention.
The food truck window is the worst possible place to split a bill. You’re standing, not sitting. You’re holding food, not a pen. People are waiting behind you. The vendor is waiting for payment. Attention that could go toward accurate calculation goes to the people waiting instead.
Groups that default to approximation still care about fairness. Accurate mental math at the food truck window is hard for most people under those conditions.
Source: Ashcraft & Kirk, “The Relationships Among Working Memory, Math Anxiety, and Performance,” Journal of Experimental Psychology: General (2001).
A practical framework
Here’s how to actually handle a multi-truck outing without cognitive overload or unfair approximations.
Designate a banker
One person handles all payments for the outing. They pay cash or card at every truck. This eliminates the “who paid what where” confusion entirely.
Photograph every order
No receipt? Take a photo of the menu board with your items circled, or screenshot your order if there’s a digital display. Memory fails. Photos don’t.
Agree on tip percentage before ordering
“We’re tipping 18% at every truck.” Done. No negotiation at each window. No one person deciding for the group.
Settle before leaving the area
While you’re still together, enter all orders into a splitting app. Assign items to people. Send requests immediately. Don’t let “I’ll pay you later” even become an option.
Round up for the banker
The person who handled all the transactions deserves a small convenience fee. Rounding their share down by a dollar or two acknowledges their coordination work.
A worked example
Let’s go back to our five friends at the food truck park. The prices are made up for this example. Here’s how the math shakes out when tracked properly.
Without proper tracking, this would have been “Taylor and Morgan, just send $20 each.” That would leave Taylor underpaying by $3.68 and Morgan overpaying by $8.12. Small amounts. But they add up over dozens of outings.
How research shaped the design
The food truck problem is really three problems: no receipts, cash payment friction, and time pressure. splitty addresses each one.
