Build note · Labor
Employee forgot to clock out, or covered a shift trade?
To tell them apart, check whether the extra hours line up with a shift somebody else was scheduled for and did not work. A covered trade leaves a matching hole in the schedule and usually leaves sales activity under the person who stayed. A forgotten clock-out leaves neither. This is the audit we built for our own locations, the four signals it uses, and the part we deliberately did not automate.
Both look identical in a timesheet
Somebody is still clocked in two hours after their shift ended. That is either wages you owe, because they covered for someone, or wages you do not, because they walked out without punching. The timesheet shows the same thing in both cases: one long punch.
The usual way this gets resolved is a manager asking around on Tuesday about something that happened Friday, which produces an answer roughly as reliable as you would expect. And the cost is rarely the wages themselves. It is the twenty minutes of reconstruction, repeated across every exception, every pay period.
The four signals that separate them
No single one of these is conclusive. Taken together they resolve most exceptions without asking anybody anything.
Signal 1
Someone else's scheduled shift
- Points to a trade:
- The long punch overlaps a shift somebody else was scheduled for and did not work
- Points to a missed punch:
- No matching gap anywhere in the schedule that night
Signal 2
The other person's punches
- Points to a trade:
- The scheduled person never clocked in at all
- Points to a missed punch:
- The scheduled person worked their shift normally
Signal 3
Where the punch ends
- Points to a trade:
- Clock-out lands near a real shift boundary, close, or another handover
- Points to a missed punch:
- Runs to the system cutoff, or to a suspiciously round hour
Signal 4
Sales activity after the shift
- Points to a trade:
- Orders, voids, or drawer activity under that person after their scheduled end
- Points to a missed punch:
- No activity at all after the scheduled end time
The part we deliberately did not automate
The audit never edits a punch. It flags the exception, shows which signals fired, and a manager decides. That was a deliberate choice, and it is the single most important thing on this page.
Any rule that automatically trims long punches will eventually cut somebody who really worked those hours. Wage corrections made silently, with no record of who decided what and why, are exactly the kind of thing that turns one annoyed employee into a wage claim. Be careful with any system that offers to clean this up for you without asking.
What the audit is genuinely good at is arriving early. Catching the exception the next morning, while the closing manager still remembers the night, is worth more than any amount of cleverness applied two weeks later.
This was a custom build
Not a standard module. The signals above depend on how your schedule, your POS, and your time clock fit together, which is different in every operation, and that is exactly the kind of gap custom builds exist to close. We built it because we kept losing manager hours to it across our own locations, not because a client filed a ticket.
If you want the standard labor tooling first, the Schedule module checks next week's labor against expected sales before you post it, and restaurant labor cost percentage covers the number itself.
Long punches, answered
How can I tell if an employee forgot to clock out or covered a shift trade?+
Check whether the extra hours line up with a shift somebody else was scheduled to work and did not. A covered trade leaves a matching hole in the schedule and usually leaves sales activity under the person who stayed. A forgotten clock-out leaves neither: the schedule is fully covered, and there is no order or drawer activity after their scheduled end time. Those two signals together resolve most cases without asking anyone.
Why does the difference matter if the hours are the same?+
Because one is wages you owe and the other is wages you do not, and correcting the wrong one is how you end up in a dispute. Paying out a forgotten clock-out quietly teaches everyone that forgetting is free. Editing down a real covered trade takes money from somebody who genuinely worked, which is far worse and much harder to walk back.
Is it safe to just auto-correct long punches?+
No. Automatically trimming any punch over a set length will eventually cut somebody who really worked those hours, and wage corrections made without a record are exactly what gets employers in trouble. The useful version flags the punch, shows the evidence, and lets a manager decide. The decision and the reason should both be recorded.
How often does this actually happen?+
In our own operations, missed clock-outs were the most common timesheet exception by a wide margin, and the overwhelming majority were genuinely forgetful rather than deliberate. That is the reason worth building for: the cost is not fraud, it is the manager time spent reconstructing what happened days later from memory.
Can Smooth Ops do this for my restaurant?+
This particular audit was a custom build rather than a standard module, because the signals depend on how your schedule, your POS, and your time clock fit together. If it is the kind of thing you need, email team@smoothopsapp.com and describe how your clock-outs go wrong today.
Tell us how your clock-outs go wrong
Describe the problem, not the solution. We will tell you straight whether it is worth building.
30 days free, no credit card