How to stop bar and restaurant revenue leakage in your hotel

Ask a Nigerian hotel owner whether their bar is losing money and most will say yes. Ask them how much, and the honest ones say they don’t know. That second answer is the real problem — not the first.

Leakage is rarely one dramatic theft. It is a slow gap between what left the store and what reached the books, opened by a process that depends on paper and memory. You cannot discipline what you cannot measure, and you cannot measure a docket that no longer exists.

Here is where the gap actually opens.

The four places it leaks

1. The docket that never arrives

A guest in A104 orders dinner and says “put it on my room.” Someone writes it on a pad. That pad now has to survive a shift change, a busy Friday night, and a walk across the property before it becomes a line on a folio.

When it doesn’t, nobody stole anything. The hotel simply served food it never billed for. At checkout the guest pays for the room, the front desk sees no restaurant charge, and everyone is satisfied — including the kitchen, which is short one cover’s worth of stock with no matching sale.

2. Stock that only exists in someone’s head

If the only record of what left the store is the same pad, the bar can never be reconciled. Two crates go out, the docket says one, and by month end the argument is about memory rather than records. Nobody can prove anything, so nobody is held to anything.

3. The drawer that three people shared

Three staff worked one till today. The float doesn’t tie out by ₦8,000. Nobody is necessarily lying — there is simply no record of who took which payment when, so the shortage belongs to everyone and therefore to no one.

4. Voids and refunds nobody sees

A void is the easiest way to make a sale disappear. If voiding costs nothing and leaves no trace, it will happen — not always dishonestly, sometimes just to fix a mistake quickly. Either way, an unlogged void is an unexplained gap.

The four controls that close it

None of these require watching your staff more closely. They change the process so the gap cannot open.

Tie the order to the guest, not the room number

A room number written on paper is a guess about which folio to bill. Linking the order to the reservation removes the guess: the charge posts onto that guest’s folio the moment it is accepted, and it can only be accepted for a guest who is actually checked in.

Then make the folio the gate. If checkout is blocked while any balance is outstanding, the dinner posted at 9pm gets settled at the desk the next morning — because the guest cannot leave otherwise.

Deplete stock at the point of sale

Selling a bottle should remove it from that outlet’s stock immediately, and reversing the sale should put it back. Once that is automatic, what left the store and what was rung up sit side by side. A gap stops being a monthly argument and becomes a specific quantity, at a specific time, on a specific shift.

Attribute every payment to a person and a shift

Every payment should carry the staff member and the shift that took it. At close, the drawer is counted against what the system says it should hold, and a shortage is flagged that night — while the people who worked it are still on the premises and still remember.

A ₦8,000 discrepancy found the same evening is a conversation. The same ₦8,000 found in next month’s audit is a write-off.

Make voids and refunds cost something to hide

Void should only be possible before money has changed hands. After payment, the correct action is a refund — and a refund should be tied to the exact payment it reverses, so refunding cash cannot also credit a folio and return the money twice.

Then log both with a person, a timestamp and a reason. Not to punish anyone; to make the record complete enough that patterns are visible.

Why the ledger has to be one ledger

Each of these controls fails if the bar and the front desk keep separate books. The room charge has to land on a folio the front desk can see. The stock movement has to come from the same sale the payment came from. The shift reconciliation has to cover both the POS drawer and the reception drawer.

That is the argument for one system rather than a PMS plus a standalone till: not that it is tidier, but that the controls only work when both sides write to the same place.

One test worth running this week: pick a Friday, count your bar stock at open and at close, and compare the difference against that night’s recorded sales. Whatever gap you find is the number you have been carrying every week without seeing it.

What this looks like in practice

Pegasus was built around these four controls specifically. The restaurant and bar POS posts room charges straight onto the guest’s folio and refuses them for anyone not in house; stock depletes on every sale; payments carry their shift; and the folio is immutable, so corrections post reversals rather than quietly rewriting history.

If you want to see it against your own outlets and your own menu, we’ll set up a demo with your figures rather than a sample hotel’s.