Search "time theft" and you will find a lot of statistics with dollar signs, nearly all published by companies that sell software to prevent it. The figures deserve more scepticism than they usually get, because the category they describe is two entirely different problems wearing one label.
One of them is real, narrow and fixable. The other is a management problem that gets worse when you treat it as a security problem.
The version that is real
Buddy punching is genuine. One employee clocks in for another who is not there, and the employer pays for hours nobody worked. It happens in shift environments — retail, hospitality, warehousing, manufacturing, care — where a shared terminal or a PIN is the only thing standing between attendance and payroll.
What makes it tractable is that it has a mechanism. Somebody has to physically enter a credential on behalf of someone else. Remove that possibility and the behaviour ends, without anyone being watched.
The fixes are unglamorous and they work. Biometric clock-in ties attendance to a person rather than a code. Geofenced mobile punching ties it to a location. Device-bound clock-in ties it to hardware the employee carries. Shift-pattern anomaly reporting catches what slips through. None of these requires monitoring what anyone does during the shift, which is why they are proportionate: they address the specific failure without collecting anything about the work itself.
If this is your problem, you want attendance tracking software, not employee monitoring. They are different categories, and buying the second to solve the first is how organisations end up surveilling an entire workforce to fix a clock-in flaw.
The version that is a category error
The other usage covers salaried knowledge workers: long lunches, personal browsing, a slow Friday afternoon, the mid-morning stretch where someone is clearly not typing.
Calling this theft imports a framing that does not fit the work. Salaried employment is not an hours-for-money exchange in the way shift work is — it is an outcomes-for-money exchange with hours as a loose proxy. Nobody bills the employer back for the Sunday evening spent worrying about a deployment, and nobody reasonable thinks they should.
More practically: the framing does not survive contact with how knowledge work actually happens. An engineer staring out of a window may be solving the problem. A designer reading unrelated material may be doing the input half of the job. A researcher whose week produced one negative result may have saved six months. Activity is such a poor proxy for value in these roles that policing it optimises for the proxy — which is productivity theatre, and it is the reliable outcome rather than an unlucky one.
Why the accusation backfires
Deploying monitoring under a theft framing has predictable effects, and they compound.
The framing is adversarial by construction, so it is received that way. An employee told the company is watching for theft understands they are suspected of theft. That is not a communication problem to be solved with better wording — it is what the framing means.
It also produces exactly the behaviour it claims to be preventing. People who believe activity is scored produce activity: the mouse stays moving, the status light stays green, the responsiveness goes up and the thinking goes down. You have not recovered the lost minutes. You have converted them into performed minutes, and lost the ability to see the difference.
And it selects against the wrong people. The staff most disturbed by being watched are usually the conscientious ones, who assume the surveillance implies a specific concern about them. The people genuinely coasting are, as a rule, the least troubled by a dashboard.
What to do instead, by situation
If you run shift or hourly work and you suspect buddy punching, fix the clock. Biometric or geofenced attendance, anomaly reporting on shift patterns, and a clear written policy. Do not extend into desktop monitoring to solve this; the mechanism is at the door, not the desk.
If you run salaried knowledge work and someone's output is inadequate, that is a performance conversation, and it is one you can have without any tooling. "This work is late and I need to understand why" is a complete management intervention. Reaching for surveillance instead is usually a way of avoiding the conversation, and it does not actually avoid it — it delays it while making it harder.
If you do not know whether output is inadequate — which is the honest position more often than managers admit — then the useful question is not who is stealing time but where the time goes. That is a systems question, and the answers are usually structural: meeting sprawl, fragmented calendars, a slow toolchain, work concentrated on two people. The cost of meetings and context switching are the two that most often turn out to be the actual leak, and neither is anyone's fault.
If hours genuinely matter because you bill clients for them, you want a time tracker, and the framing is billing accuracy rather than theft.
The honest version of the concern
None of this is an argument that everyone is working hard. Some people coast. Distributed work made coasting easier to sustain and harder to see, and it is reasonable for a manager to want a clearer picture.
The useful reframe is that you are looking for a signal, not a culprit. A team where several people show sustained low engagement has a problem, and it is far more often a management, workload or role-fit problem than a character problem — the quiet quitting data is fairly clear that disengagement tracks workload and boundary drift rather than generational attitude. Aggregate patterns will show you that. Per-person surveillance will show you something that looks like an answer and mostly is not.
Where ProdView fits
ProdView is built for the systems question rather than the culprit question. It measures activity and app metadata to show focus time, meeting load, capacity distribution and working patterns, aggregate-first, never screen content or keystrokes, with employees on the same dashboard as their manager. $4.99/user/month, three seats free.
When not to pick us: you have a genuine buddy-punching problem. We do not do biometric or geofenced clock-in and would not solve it — Jibble and similar attendance tools do, and that is the right purchase.
Try before you commit
If the suspicion is really about where the week goes, three seats are free forever and a month of data will settle it more honestly than an argument will. The ROI calculator is free and needs no signup if you want to price the process problems the data usually surfaces.