There are three tools people use to settle a group dinner. None of them were built for a group dinner.
This is the actual problem. Not that bill splitting is hard. Not that math is annoying. The problem is that the people who designed the tools we use weren't thinking about you sitting at a restaurant in San Diego on a Thursday night with five friends and a $284 bill. They were thinking about something else.
Here's what each one was actually built for, and where each one falls apart.
Splitwise
Splitwise was built for roommates.
The product is brilliant for that. You and your two roommates split rent, utilities, groceries, the new couch. Splitwise tracks who paid what and shows running balances over months. "You owe Jake $74" updates automatically as expenses get logged. At the end of the month, you settle up. It works because everyone in the group is the same five or six people, recurring forever.
The moment you take Splitwise to a restaurant, three things break.
First, you have to add everyone to the group before you can split anything. At a dinner with five friends, that's five email addresses, five names, five "wait what's your last name" moments. By the time the group is set up, the food has arrived.
Second, Splitwise doesn't actually move money. After you've calculated who owes what, everyone has to leave Splitwise, open Venmo, manually enter the amount, and pay. The app's job ends right where the friction begins.
Third, the running balance model — the whole thing that makes Splitwise work for roommates — actively hurts you for one-time dinners. You don't want to "owe Sarah $23" floating in an app for the next month. You want to be square before you leave the table. Splitwise's strength is its weakness in this context.
Venmo
Venmo was built for paying one person at a time.
You owe your friend $40 for the concert ticket? Venmo. You're paying back your sister for the gift? Venmo. The app does this beautifully — fast, free, social, frictionless.
What Venmo cannot do is split a bill among five people. There is no "this bill, divided by these line items" feature. You can request money from multiple people in a group, but the request is for a flat amount you typed in. Venmo doesn't know what was on the receipt. It doesn't know about tax or tip. It doesn't know who had the wine.
So Venmo gets used at the end of a long process that happens somewhere else. Someone does the math (in their head, in a calculator, in Splitwise), tells everyone what they owe, and then people Venmo. Venmo is the last 30 seconds of a 10-minute process.
It's also the moment where the IOU starts. Once people walk away from the table, payment requests get ignored, forgotten, deferred. You sent five Venmo requests. By tomorrow, three are paid. By next week, you're chasing one. The math was right; the human follow-through was the problem.
The Receipt Itself
This is the most common solution and the worst one.
The waiter brings the check, someone picks it up, and the table starts negotiating. "I'll pay for ours and you Venmo me." "Just split it five ways." "Wait, did the appetizer get on this bill or the other one?" Twenty minutes later, the receipt is folded into someone's wallet, and a vague chain of debts has been verbally established.
This works at a coffee shop. It does not work at a $300 dinner with two people who didn't drink, one person who ordered the most expensive thing on the menu, and a service charge that nobody noticed.
Research on cognitive load makes this concrete: George Miller's Psychological Review paper showed working memory holds about seven items, plus or minus two. A restaurant receipt with fifteen line items, tax, tip, and five people to assign them to is well beyond what any one person can hold in their head while having a conversation.
So things get fudged. The math gets approximated. Someone always pays slightly too much, someone always pays slightly too little, and over years of dinners with friends, those small inaccuracies add up to real money — and small resentments.
What's Actually Missing
The honest answer is: a tool built for the moment.
Not for ongoing roommate ledgers. Not for one-on-one transfers. Not for memory-based negotiations at the table. For the specific moment when a group of people has just finished eating and needs to pay their exact shares quickly without anyone doing math.
That tool needs four things, and none of the existing tools have all four:
- It has to read the receipt automatically. Nobody types in 15 line items.
- It has to work without anyone signing up. The host shouldn't have to onboard their friends.
- It has to handle proportional tax and tip. Even splits are unfair; the person who ordered the steak should pay more tax on the steak.
- It has to actually move the money, on the spot, with one tap.
Splitwise has #3. Venmo has #4. The receipt has neither. The combination — receipt scan + shareable link + proportional math + Venmo deep-link — has been waiting to be built.
That's what Divvyme is.
The host scans the receipt. The link goes in the group chat. Each guest taps what they ordered, on their own phone. Venmo opens with the exact amount pre-filled. No accounts, no math, no "who had the wine," no IOUs sitting in someone's app for a month.
The bill that used to take ten minutes takes about ninety seconds.
Splitwise is great. Use it for your roommates. Venmo is great. Use it to pay your sister. But for the moment between "check please" and "everyone's even" — at a dinner, with friends, on a Thursday night — those tools were never designed for what you're trying to do.
Now there's one that was.
Miller, G.A. "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information." Psychological Review, 1956.
Scan a receipt, share a link, everyone pays their exact share. Under two minutes. No download. No account. Free.