Working out who owes who after a trip, properly
· 4 min read
The reason group trip balances go unsettled is almost never the amount. It is that the task looks enormous.
Six people who all paid for different things appear to owe each other in every direction. Counted pairwise that is fifteen possible debts, and fifteen transfers is not something anybody is going to sit down and complete. So it does not happen, and four months later somebody makes a joke about it in the group chat.
The good news is that fifteen is an illusion. It nearly always reduces to three or four. Here is the method, worked through, so you can either do it yourself or check that your app is doing it properly.
Step one: stop thinking in pairs
The instinct is to track "Rahul owes Priya ₹400, Priya owes Ananya ₹250" and so on. That is the trap. Pairwise debts multiply — n people gives you n(n−1)/2 possible pairs, so six people is fifteen and eight people is twenty-eight.
Instead, give each person a single number: what they paid, minus what they owed. Positive means the group owes them. Negative means they owe the group.
Step two: work out the net positions
Six friends, four nights. Total spend ₹72,000, so an even share is ₹12,000 each.
| Person | Paid | Share | Net |
|---|---|---|---|
| Priya | ₹34,000 (villa) | ₹12,000 | +₹22,000 |
| Rahul | ₹16,000 (dinners) | ₹12,000 | +₹4,000 |
| Ananya | ₹14,000 (car, fuel) | ₹12,000 | +₹2,000 |
| Vikram | ₹5,000 | ₹12,000 | −₹7,000 |
| Rohit | ₹3,000 | ₹12,000 | −₹9,000 |
| Meera | ₹0 | ₹12,000 | −₹12,000 |
Check the positives and negatives cancel. +22,000 +4,000 +2,000 = +28,000. −7,000 −9,000 −12,000 = −28,000. They match.
If they do not match, do not go any further — an expense is missing, double-counted, or someone's share is wrong. This check catches nearly every data-entry error, and it is the step people skip.
Step three: match biggest to biggest
Repeatedly pair the largest debtor with the largest creditor and transfer the smaller of the two amounts. One of them clears each time, so it terminates quickly.
Meera owes ₹12,000. Priya is owed ₹22,000. Meera pays Priya ₹12,000. Meera is clear; Priya is now owed ₹10,000.
Rohit owes ₹9,000. Priya is owed ₹10,000. Rohit pays Priya ₹9,000. Rohit is clear; Priya is now owed ₹1,000.
Vikram owes ₹7,000. Priya is owed ₹1,000. Vikram pays Priya ₹1,000. Priya is clear; Vikram still owes ₹6,000.
Vikram owes ₹6,000. Rahul is owed ₹4,000. Vikram pays Rahul ₹4,000. Rahul is clear; Vikram still owes ₹2,000.
Vikram owes ₹2,000. Ananya is owed ₹2,000. Vikram pays Ananya ₹2,000. Both clear.
Done. Five transfers instead of fifteen, and notice that Rahul and Ananya — who were owed money — make no payments at all. They just receive.
The general result is that you always need at most one fewer transfer than there are people. Six people, five transfers maximum, usually fewer.
Why this is worth doing
Compare the two lists you could post in the group chat.
Fifteen lines of "X owes Y ₹340" is a document people close. Five lines is a task people finish in ten minutes.
That is the entire difference between a settled trip and a balance that sits there. The arithmetic is not the point — the point is producing a list short enough that somebody actually starts.
Doing it with uneven shares
The method is identical when shares are not equal. Two people skipped the day trip, one person had the big room — just compute each person's actual share of each expense, sum it, and subtract from what they paid. You still end up with one net number per person, and steps two and three are unchanged.
The only real risk is arithmetic errors at the share stage, which is what the sum-to-zero check catches.
Doing it automatically
This is exactly the sort of thing worth handing to software, not because the maths is hard but because doing it by hand for six people takes ten minutes and gets it slightly wrong often enough to matter — and a slightly wrong ledger is worse than none, because now there is an argument.
TripSynk computes the reduced set as you go, so the short list is there when you want it rather than being a thing someone has to sit down and do. You still make the transfers yourself in your UPI or banking app — no splitting app in India moves money — and the app records them so the balances clear.
Whatever you use, the thing to check is that it reduces. An app that shows you a pairwise list has done the tracking and left you the part that actually stops people.