New · OptSolution School, OPTimes, OPTGateway, OPTrace: discover our latest launches

Fraud prevention

Ghost workers: the real problem is not clocking in, it is staff enrolled nowhere

A time clock counts the people who turn up. It says nothing about those who draw a salary without ever passing in front of it, and that is where the wage bill leaks.

Staff queue at an enrolment desk where an HR officer records their fingerprints.
AI-generated image (Higgsfield)

What the time clock cannot see

A time clock records the people who come and stand in front of it. That is its job, and it does it well. But a ghost worker, by definition, never turns up. They do not show as absent: they do not show at all. Meanwhile, they are on the payroll and draw a salary every month.

That is why an attendance report can look excellent while the wage bill keeps leaking. The report measures how reliably enrolled staff turn up. It says nothing about the people who are not enrolled.

So the real problem is not clocking in. It is knowing, name by name, who is not enrolled anywhere.

What a census reveals

On one deployment of OPTimes, our platform that links biometric attendance to payroll, the workforce stood at 7,724 staff. The first review showed that 6,241 of them were not enrolled on any terminal, and that 232 were in the system with no fingerprint, no face and no card.

In other words, at that point, more than four in five staff had no way of proving their attendance through the system.

This figure needs reading honestly. It does not mean there were 6,241 ghost workers. Most of these people exist and do their jobs: the enrolment drive had not yet reached their site, their entrance had no terminal, or they were on assignment, on leave or seconded elsewhere. But until they are enrolled, the organisation cannot tell a real employee who is absent from one who does not exist. That grey zone is where abuse thrives.

The 232 records with no biometric data at all are the first list to look at.

Why the attendance rate is not enough

Many organisations track a single indicator: the attendance rate. Calculated only on enrolled staff, it can look very healthy while the payroll is paying people nobody has ever seen clock in.

A useful dashboard counts three groups, not one:

  • enrolled staff, whose attendance is measured;
  • unenrolled staff, whose attendance cannot be measured;
  • staff with no biometrics at all, whose very identity the system has never checked.

Until the last two groups are close to zero, the attendance rate describes part of the workforce, not the workforce.

What a biometric census changes

A well-run census turns an administrative list into a list of real people. Each member of staff comes forward, is identified and enrolled. From then on:

  • every salary matches someone who has been seen and recorded;
  • every absence is a genuine absence, not a blank;
  • payroll can be calculated from recorded attendance, with nothing keyed in twice between the terminal and the payslip.

That last point is what closes the gap for good. When attendance and payroll live in two separate systems, with a spreadsheet copied across every month, the hole opens again.

The method, step by step

  1. Measure. Produce the named list of unenrolled staff and of those with no biometrics. A platform that cannot produce that list cannot help you take back control.
  2. Get departments to confirm. Each head of department states who actually reports to them, where they work and since when.
  3. Call people in for enrolment. A schedule for each site, with a clear deadline, and a separate route for staff on assignment, on leave or based in the provinces.
  4. Deal with the remaining cases. Anyone who does not come forward despite reminders has their file reviewed, in line with the organisation's rules and applicable labour law.
  5. Keep the list current. Every new starter is enrolled before their first salary. Every leaver is removed on the day they go.

Mistakes to avoid

  • Assuming fraud. Treating every unenrolled person as a suspect would be unfair and wrong. The aim is to verify, not to accuse.
  • A one-off census with no follow-through. A clean list decays within months if joiners and leavers do not go through the same process.
  • Simply buying time clocks. Terminals without enrolment tracking only measure the people who had already decided to turn up.

You cannot check the attendance of someone you have never identified.

If you want to know how many of your staff are not enrolled anywhere, that is where we start. Tell us the size of your workforce and the number of sites.

← All articles

A similar project on your side? Let's talk.

Ghost workers: the real problem is not clocking in, it is staff enrolled nowhere — Enywork SARL