Working Time Records

Without Buying Software

This is not legal advice.

The duty does not require a purchase, and for a small employer with regular hours the entire implementation fits on one page. Here it is. For an additional implementation-oriented example, see this reference.

Reviewed August 9, 2026.

What you need

A form, per employee, per month. Four columns: date, start, end, breaks. For additional workplace and technology context, see Destatis.

A routine. The entry is made on the day, by the employee or by whoever is present.

A check. Weekly, by a named person, confirming records exist and are plausible.

A place. Where completed sheets live, in order, for the retention period.

And a rule for corrections. Struck through, initialled, dated. Never overwritten.

That is the whole system, and it meets objective, reliable and accessible if the five parts are actually done.

The form

Employee: ___________          Month: ________

 Date  | Start | End   | Breaks | Signature
-------|-------|-------|--------|----------
  01   | 08:15 | 17:00 | 00:45  |
  02   | 08:00 | 16:30 | 00:30  |

Duration is derivable from start, end and breaks, which is what the duty needs.

A signature column does the accessibility work: the employee has seen and acknowledged their own record.

And a corrections line at the foot, for anything struck through, with a reason.

The routine

Same day. The draft proposes it as a rule and it is the difference between recording and reconstructing.

One place. A folder at the site, a shared drive, a tray. Not several people's individual arrangements.

And a habit attached to something that already happens — end of shift, locking up, the last task of the day. A recording step with no trigger gets skipped exactly on the busy days.

The check

Weekly, ten minutes, by a named person.

Do records exist for everybody who worked? Gaps are what an inspection notices and what a dispute exploits.

Is anything implausible? Fourteen-hour days, missing breaks, identical times every day for a month.

And chase the gaps within the week, while people remember. Reconstruction later is worse than an acknowledged gap.

The check is the part that gets omitted, and it is what converts a pile of forms into a system.

Where this stops working

Honestly, because the page is arguing for the cheap option.

Shift work and variable hours. Reconciliation becomes the job.

Multiple sites. One place stops being one place.

Payroll dependence. Manual transcription introduces errors that cost more than a licence.

Field staff and drivers, where a central paper record is impractical.

And above roughly ten to fifteen employees, where the weekly check needs somebody to own it as a task rather than as a habit.

At any of those points buy something — because it suits the operation, which is a better reason than a deadline.

The spreadsheet version

Between paper and a product, and frequently the right answer.

Same four columns, one sheet per employee, one file per year.

Advantages: legible, searchable, sums itself, and exports.

The thing to get right: a change history. A spreadsheet silently overwritten fails "reliable" exactly as an erased pencil entry does. Version history in a shared drive gives you this for nothing, and it is the single feature that makes the spreadsheet defensible.

The short version