How to eliminate buddy punching in a security company

Buddy punching is when one employee clocks in or out on behalf of another who is not actually there. The timesheet says the shift was worked. Payroll pays for it. Nobody was on post.

In most industries this is a payroll leak. In contract security it is also a contract-integrity problem, because the thing being sold is presence. A shift that was billed but not worked is not just an overpayment. It is a promise to a client that was quietly broken.

Almost every structural feature of guarding works against detection:

• Guards work alone, so there is no colleague who would notice an empty post.
• Shifts run overnight, when no one from the office or the client is on site.
• Sites are geographically scattered, so supervision means driving, not walking down a corridor.
• The work is intentionally uneventful, so a quiet shift and an unworked shift produce identical paperwork.

The result is that discovery is usually accidental and late. A client notices an unlocked gate, or asks for footage, and the question becomes whether anyone was there at all. By that point the problem is not the payroll. It is the account.

Each of these is weak alone and strong in combination. The goal is not a perfect system; it is making the dishonest path harder than simply working the shift.

1. 1Tie the clock-in to a place. A geofence means the clock-in has to physically happen at the site, which removes the phone-from-home case entirely.
2. 2Tie the clock-in to a person. A photo captured at clock-in ties the record to a face rather than a password that can be shared in a text message.
3. 3Require presence to continue. Patrol checkpoints during the shift mean showing up once is not enough, because the guard has to still be there at 02:00.
4. 4Surface the exceptions. A late or missing clock-in that nobody looks at is not a control. Someone has to see it while the shift is still running.

• Shared site tablets or kiosks: anyone standing there can clock in anyone, which is the original problem with extra hardware.
• PIN codes: a PIN is a piece of text, and text can be sent.
• Call-in check-ins to a dispatcher: caller ID says nothing about location, and it costs someone a night of answering phones.
• Retroactive timesheet review: it finds the discrepancy weeks later, long after the client would have.

Guards generally do not object to verified clock-in. They object to being told, mid-contract and with no explanation, that they are being tracked. The framing matters and so does the scope.

1. 1Explain what is collected and when. For location, that means during scheduled shifts only, not off the clock.
2. 2Put it in writing before it goes live, and make sure supervisors can answer questions about it.
3. 3Start with one site, confirm the geofence radius is not producing false rejections, then expand.
4. 4Tune per site. A downtown high-rise and an open industrial yard do not want the same radius.
5. 5Give guards a working fallback for the day a phone dies, so the control does not push honest people into workarounds.

• Geofenced clock-in: guards can only clock in from inside the site boundary, and GPS coordinates are stored on the record.
• Selfie verification at clock-in, stored with timestamp and device information.
• Patrol checkpoints so coverage has to continue through the shift, not just start.
• Location collection limited to active, scheduled shifts, never off the clock.