Who Does the Recording

This is not legal advice.

The duty says records must exist. It does not say whose hands make them, and all three usual arrangements are permitted. For an additional implementation-oriented example, see employee PC activity tracking.

What does not move is the responsibility. Completeness and correctness stay with the employer regardless of who types. For additional workplace and technology context, see Gesetze im Internet.

Reviewed August 9, 2026.

Employee self-recording

The most common arrangement and the natural fit for flexible work.

Good at: accuracy about what actually happened, because the person was there. Low friction. Works with trust-based arrangements without changing them.

Fails at: consistency. Entries get skipped on busy days, which are the days worth recording, and the gaps appear weeks later.

Requires: a named person who checks that records exist and chases gaps. Without that check it is delegation of the task and retention of the liability.

Manager recording

Somebody else enters the times for a team.

Good at: consistency, and it suits shift work where the shift pattern is the record.

Fails at: accuracy about departures from the plan. A manager recording scheduled hours is recording a schedule rather than a record, and the overtime that the duty exists to make visible is exactly what goes missing.

Requires: employee visibility. A record the employee cannot see is the arrangement that produces disputes, and it is worth avoiding even where it is lawful.

System recording

Badge, terminal, app, or login-based capture.

Good at: consistency and objectivity. Nothing is forgotten and nothing is estimated.

Fails at: the boundary. A badge records arrival at the building, not the start of work; a login records the machine, not the person. The gap between the two is small and it is systematic, which is worse than random.

And it captures more than the duty requires if configured carelessly — which is where a compliance system becomes a monitoring one.

Requires: a correction route that is easy to use, because automatic capture is wrong more often than people expect.

The hybrid that usually works

System or employee capture, with a weekly human check.

The capture handles volume and consistency; the check handles gaps, implausible entries and corrections.

Ten minutes a week for a small employer, and it is the part that converts records into a system rather than a pile.

The decision, by situation

Regular hours, small team, one site: employee self-recording, weekly check. No purchase needed.

Shift work: system capture against the roster, with departures recorded and visible.

Field staff and drivers: employee capture on a device, because a central record is impractical.

And flexible or remote work: employee self-recording, because nobody else knows.

The one rule across all of them

The employee can see their own record.

It is required by the accessibility limb, it is what makes self-recording meaningful, and it is the single feature that keeps manager and system recording honest. An arrangement where the employee cannot check what was written about their hours has failed the point of the exercise, whatever else it satisfies.

The short version

• All three arrangements are permitted, and responsibility for completeness stays with the employer regardless
• Employee recording is accurate and inconsistent — it needs a named person chasing gaps
• Manager recording is consistent and drifts toward recording the schedule, which loses the overtime the duty exists to reveal
• System recording is objective and misses the boundary between arriving and starting, systematically rather than randomly
• The hybrid that works: capture plus a ten-minute weekly human check
• One rule across all of them — the employee can see their own record