Punch Time Is Not Arrival Time

The first thing a new clock-in system produces is an accurate-looking record of who was late. That record is less accurate than it looks.

What sits between arriving and punching

Getting a fix, which indoors can take a while or fail entirely.

Retrying after a refusal.

Walking from the entrance to wherever the punch is possible, if coverage is poor at the door.

Queueing, at a shared terminal at shift change.

Opening the app, waiting for it to load, tapping.

None of this is the person being late, and all of it lands in the same column.

The rigidity trap

A precise timestamp makes a one-minute rule feel enforceable.

Which produces: punching from the car park to be safe, asking colleagues to punch, resentment about minutes, and people standing in doorways trying to get a fix instead of starting work.

The system then records the behaviour it caused, and the behaviour looks like a workforce problem.

Slack in the policy removes most of it and costs almost nothing.

What the data can honestly support

Patterns over weeks, at site level: a site where everyone arrives at the same slightly late minute has a bus timetable, not a discipline problem.

A person whose pattern changed, which is a conversation rather than a finding.

Aggregate punctuality trends, as context.

Not: a minute-level record used as evidence against an individual, because the minutes contain everything above.

Before any lateness conversation

Check the site's refusal rate.

Check the accuracy figures on that person's punches.

Check whether the entrance has coverage, by standing there.

Check the rota was published in time, because late rotas produce late arrivals.

Most lateness patterns resolve at one of those four, and the conversation is then about a bus or a rota rather than about the person.

Where to put the boundary

At the perimeter, and accept that the walk to the workstation is paid.

Or at the workstation, and accept that the walk is unpaid, which in several jurisdictions is contested and worth checking.

Decide deliberately and say which, because an undecided boundary means the answer is whatever the coverage happens to allow.

The test of the policy

Does anyone punch from outside to be safe?

If yes, the policy is too rigid or the coverage is too poor, and either way the record has stopped describing arrival times.

Ask people. They will tell you, and the answer is more reliable than the data at that point.

Check four things first

Before any lateness conversation.

The site's refusal rate.

The accuracy figures on that person's punches.

Whether the entrance has coverage, by standing there.

Whether the rota was published in time.

Most patterns resolve at one of those four, and the conversation is then about a bus or a rota rather than about the person.

Connect policy to configuration

The practical choices behind this note can be compared with the service-desk workflow. Keep the written purpose in control and enable only the data needed for it.

Independent reference

For an external point of reference, see the ACAS website. Its published guidance provides useful context for testing the assumptions in this note.