Ask any owner of a forty-room hotel whether their cash balances and you will get a small pause before the answer. Not because anything dramatic is going on — because the honest answer is "mostly", and because the gap between mostly and exactly has never been worth the argument it would take to close.
This is not really a cash problem. It is a question of attribution: the money is short, and there is no way to say whose shift it went missing on. Fix that and most of the shortage quietly disappears, because nearly all of it was never theft in the first place.
Where the money actually goes
In the great majority of properties, a short till is the sum of entirely reasonable things nobody wrote down:
- A guest settles part of a bill in cash at 11pm and the rest goes on a card in the morning.
- Somebody pays the vegetable supplier out of the till because the bank is shut.
- A refund is handed back from the drawer without a receipt.
- An advance is taken for a booking and put in the drawer with a mental note.
- A staff member borrows two hundred and puts it back on Friday. Usually.
Each is small. None is dishonest. And at the end of the week the count is out by an amount that nobody can decompose, so it gets written off and the same thing happens next week.
The dangerous part is what that habit costs you: once a property accepts that the till is approximately right, actual theft becomes invisible. It hides inside the noise you have already agreed to ignore.
A count that names the person who held the till
Open with a float, watch the expected figure build through the shift, count at handover — and the difference is recorded against a name, with a note.
Start free
Why counting once a day cannot work
Most properties count at the end of the day. It feels rigorous and it tells you almost nothing.
A daily count produces one number for a twenty-four hour period in which three or four different people held the drawer. If it is short by 800, that figure belongs to everybody and therefore to nobody. You cannot act on it. You can only mention it at a staff meeting, watch everyone look at the floor, and repeat the exercise next month.
A shift count produces the same number against one person and a window of a few hours. That is a conversation with a beginning: what happened on Tuesday afternoon? And more often than not the answer is completely innocent — the supplier, the refund, the advance — which is exactly the point. You want the innocent explanations surfaced, because that is what leaves the unexplained ones visible.
The four things that make a shift count work
1. The drawer belongs to a person, not to the desk. This is the whole mechanism. Named sign-in is what turns a count into accountability. In Zitlin that means staff sign-in with a 6-digit PIN — each person taps their name and punches six digits on the shared terminal. No email, no password, nothing that works from home. The device stays signed in as the property; the PIN just says who is standing at it.
2. One physical drawer per person. Software cannot save you here. If two cashiers share a till, the difference at the end belongs to neither of them and the whole exercise collapses. Zitlin allows several drawers open at once — a front desk and a restaurant till, counted separately, each answerable to whoever holds it — precisely so you never have to share.
3. The expected figure has to be visible and explicable. The cashier needs to see how the system arrived at its number: the opening float, the cash settlements taken, the cash expenses paid out. Otherwise they cannot tell a mistake from a discrepancy, and they will assume the software is wrong.
4. The count is recorded, not negotiated. When the figures differ, the system should take the number, require a note, and close the shift. Software that argues with the count teaches staff to enter whatever it will accept — and then you have a tidy report and no information at all.
What to do on Monday morning
You do not need new software to start. You need three decisions:
- Define your shifts — Morning, Afternoon, Night, or whatever you call them, with a start time for each. Fixed names, not typed ones: "Morning", "morning" and one bad day's typo become three shifts no report can reassemble.
- One drawer, one person, every handover. Counted at the change, not at the end of the day.
- A note is compulsory whenever it doesn't balance. Not a punishment — a sentence. "Paid 450 to the laundry, receipt in the file" is the whole point of the exercise.
Do that on paper for a fortnight and you will already know more than you did. Most owners discover the pattern is boring: one shift, one recurring cause, fixed in an afternoon.
Close the Shift Before You Hand Over the Keys
How Zitlin does it
Cash drawer and shift handover is part of Zitlin Pro. The shape of it:
- The person taking the desk opens a drawer with a counted float, against their own name.
- Through the shift, Zitlin builds the expected figure — cash settlements in, cash expenses out — so the cashier can see where it came from.
- At handover they count what is physically there. Zitlin records the difference against them and requires a note when the two disagree.
- Card, UPI and transfer totals are shown but never reconciled — no drawer moves when a guest taps a card, so they carry no short or over. They are there to tick off against your terminal's batch.
- Shift history filters straight to the counts that did not balance, so three weeks later you can still find the short shift and the reason someone gave for it.
- Where several people work one shift, the shift report rolls their drawers into a single document: what the property took, and who moved it.
Two things worth knowing before you plan around it. The drawer requires staff sign-in to be switched on first — a drawer belongs to a person, and without named sign-in there is nobody to hold answerable. And your shift list is set up by Zitlin once when the feature is enabled, so tell your contact how many shifts you run and when they start.
The part that isn't about money
The reason to do this is not to catch anybody. It is that a property where every till is counted against a name is a property where the question "who did this?" has an answer — for voids, discounts, rate overrides and refunds as much as for cash. That changes behaviour long before it catches anyone, and it makes the rare genuine problem findable instead of invisible.