Clocking in early: the labor cost nobody calls theft
· payroll, time & attendance, labor cost
At a small hotel the expensive timekeeping problem is almost never the one anybody argues about. It is nine minutes — the gap between arriving and starting, paid, six days a week, by everybody. Across six hourly staff that is about 280 hours a year, and nobody ever raised it, because no single instance was worth raising. The reason it persists is not that people are stealing. It is that nobody ever said when the shift starts.
What clocking in early actually costs, at six people and nine minutes
An invented property: a 46-room motel, six hourly staff clocking in on a kiosk at the back office, each working six shifts a week, placeholder wage of $19 an hour.
| Shifts a week | 6 staff × 6 = 36 |
| Average early clock-in | 9 minutes |
| A week | 324 minutes = 5.4 hours |
| A year | 281 hours |
| At $19 | $5,335 |
Add the other end of the shift. A clock-out at 5:06 for a shift that ended at 5:00 is another 6 minutes; on the same 36 shifts that is 187 hours and $3,557 a year. Nobody at this property is doing anything wrong, and the two habits together come to 468 paid hours nobody worked — about nine hours every week, $8,892 a year.
The number is also worse than it looks in one specific way. If any of those staff are near a daily or weekly overtime threshold, the drifted minutes are the ones that cross it — so a share of that total is paid at a premium rate rather than straight time. A weekly threshold applies federally, and some states add a daily one on top of it — which states, and what they actually require, is its own subject. Those rules differ and change, so which of them bites you is a question for your own counsel rather than for a blog post.
None of this is theft, and the word costs you money
“Time theft” is the term the category uses, and at a property this size it is usually the wrong description of what is happening. A housekeeper who arrives at 7:51 for an 8:00 start and clocks in because she is standing next to the clock is not taking anything. She is doing the obvious thing with a device nobody gave her a rule for.
The framing matters commercially, not just morally. If you open this conversation by implying dishonesty, you will get a defensive, resentful staff and you will still have the nine minutes, because the cause was never dishonesty. Small properties run on people who will cover a shift at two hours’ notice, and that is not what you want to spend to recover nine minutes.
The one genuine dishonesty case — somebody punching in for a colleague who is not there — is a real and separate problem, and it has its own fixes. Do not use those fixes on this problem. They address a different thing.
Why it happens: nobody wrote the rule down
Ask at most small hotels when the shift starts and you will get three answers: when you arrive, when you clock in, and when the outgoing person leaves. Those are different moments and they can be twenty minutes apart.
Drift grows in that gap. It also grows wherever arriving and starting are the same physical act — if the clock is the first thing you walk past, clocking in is what arriving looks like. And it compounds quietly, because nine minutes never triggers anything. Nobody queries a timesheet over nine minutes, so the number is never examined until somebody adds up a year of it.
What actually reduces it
Three things, cheapest first.
State the rule, once, in plain language. “Clock in at your shift start, not when you arrive. If you are asked to start early, that is a shift change and it is paid.” That costs nothing, it converts an ambiguity into an expectation, and it does so without accusing anyone.
Make arriving and clocking in separate acts. If the clock sits inside the area where work begins rather than at the door, the two stop being the same gesture. This is a furniture decision more than a software one, and it is free.
Look at the day, not the period. A drift of nine minutes is invisible in a month’s total and obvious in a day’s. Punches at this kind of property need approving before they become payroll, and ProfitClock leaves anything unapproved sitting in the daily labor report as an exception rather than letting it quietly become a number — which means the conversation happens the same week, while it is still a small conversation. Start a free trial and watch one week of starts against one week of shifts.
Geofenced clock-in is worth a sentence here because it is frequently sold as the fix for this and is not. It tells you the punch happened at the property; it says nothing about whether the shift had started. It solves a different failure.
The thing not to do: quietly edit the punch
The tempting shortcut is to shave the punch back to 8:00 at approval time and say nothing. Do not. Three reasons, and the third is the one that leaves you unable to explain the record later.
- It hides a real scheduling signal from you — you will keep paying for the drift you just erased from your own view of it.
- It is invisible to the person it affects, so the behaviour never changes.
- Adjusting a time record after the fact, without a trace of who changed what and why, is a records problem. Whether and how recorded time may be rounded or adjusted is a wage-and-hour question with a real answer and a genuinely complicated one; check your own state’s rules and take advice — the questions worth putting to a vendor about it are a better starting point than a blog post’s opinion. The operational rule that holds regardless: an edit should be visible, attributed, and leave the original readable. ProfitClock keeps punch edits as an append-only history for exactly that reason.
Where the honest limits are
Three places this argument does not reach.
- If people clock in early because the work genuinely starts then, the schedule is wrong and the staff are right. A 7:45 arrival to strip the breakfast room before an 8:00 “start” is unpaid work being done, and the fix is to move the shift, not the rule.
- Breaks are their own subject. Unpaid breaks that are not clocked cannot be measured, and what you may require varies by jurisdiction in ways that are not a rounding detail. We are deliberately not answering that here.
- Nine minutes is this invented property’s number, not a benchmark. Yours might be two, and then the whole exercise is worth $1,186 and probably not worth a meeting. The arithmetic above is the useful part: shifts a week × average minutes ÷ 60 × your rate × 52. Run it before deciding it matters.
One more thing worth measuring while you are in there: the day boundary. If your system decides which day a punch belongs to in UTC rather than at the hotel, some of these minutes are landing on the wrong day entirely — and in a state with a daily eight-hour threshold, a ten-hour evening shift split across two dates loses the two overtime hours it actually earned.
Start seeing what labor actually costs
Set up your property, put staff on the clock, and read tomorrow morning’s labor number.
14-day free trial · No credit card · Cancel anytime