Employee time tracking without buying a system for it

Employee time tracking is one of those practices that gets adopted before anyone has said what it is for. Somebody suggests it, a tool gets bought, everyone starts filling in a grid, and six months later nobody looks at the data. The tracking that survives is the tracking that feeds a decision: an hourly payroll, an invoice to a client, or a genuine question about where capacity goes.

So before choosing anything, write the question down. "How many hours did Ana work in July" is a payroll question and needs daily totals. "How much of this month went to Acme" is a billing question and needs hours by client. "Why does every project overrun" is a planning question and needs hours by activity, sampled for a few weeks, not tracked forever.

What employee time tracking has to capture

• Who, and for which period — day and week, since almost every downstream use is weekly.
• Hours worked, in the unit you will actually use. Quarter hours are precise enough for payroll and billing; minute-level precision creates false accuracy.
• What the hours went to: client, project, or activity. One dimension, not three — multi-dimensional coding is where compliance collapses.
• Overtime, separated, because it is usually paid differently and almost always reported separately.
• Absence: holiday, sick, unpaid. If these are not in the same record, someone reconciles two systems by hand every month.
• Approval — a manager confirming the week — and the date it was confirmed.
• A lock. Once a period is approved and paid, the record for it stops being editable.

Check the rules where you operate before designing anything. In the EU, working-time records are a legal obligation, not a management preference, after the 2019 CJEU ruling; the UK, the US and others have their own record-keeping and retention rules for hourly staff. The design below is generic — the retention period and what must be recorded are not.

A spreadsheet-first setup that works

1. One row per person per day, not one sheet per person. Rows are what let you filter, pivot and total; separate sheets are what force copy-paste at month end.
2. Columns: date, person, project or client, hours, category (normal, overtime, holiday, sick), note. Seven columns handle most companies.
3. Use a dropdown list for project and category so entries stay comparable. Free text here is what makes the data unusable later.
4. Add a weekly pivot: hours by person, hours by project. This is the whole reporting layer for a team under about thirty people.
5. Approve weekly. A manager checks the week, marks it approved, and after that the rows are protected.
6. Export the approved period as CSV or XLS for payroll and keep the file — the export is the record, and it needs to survive later edits to the sheet.
7. Review the pivot monthly against what you expected. If nobody looks at it, stop collecting it: tracking that feeds nothing is a tax on everyone's Friday.

Getting people to fill it in

Compliance is a design problem. Daily entry takes under a minute if the sheet is open and the dropdowns are short; weekly reconstruction takes twenty minutes and produces fiction. The fix is not reminders, it is making entry small: fewer projects on the list, fewer categories, no mandatory description field. If your list of projects does not fit on one screen, the list is the problem.

Be explicit about what the data is used for, too. People fill in timesheets accurately when they know it drives an invoice or their pay, and vaguely when they suspect it is surveillance. If the reason is capacity planning, say so, and share the pivot with the team — the fastest way to get honest hours is to give the people entering them something back.

Doing it in Ettex

Ettex Sheets is a reasonable home for this: import an existing XLSX with formulas intact, use dropdown validation for project and category, conditional formatting to flag missing days or over-long weeks, pivots for hours by person and by project, and cell comments with threads when a manager queries a line. Everyone edits the same sheet live rather than mailing versions around, version history means an altered week is traceable, and export to CSV or XLS hands payroll the file it wants. If entries arrive from people who should not see the whole sheet, collect them through Ettex Forms instead — each submission timestamped in one inbox — and pull the submissions into the sheet weekly.

The boundary, plainly: Ettex has no stopwatch, no automatic capture of what you were doing, no GPS or screenshots, no payroll engine and no billing integration. This is manual entry into a well-structured sheet with review and export. That covers a large share of small companies; if you need running timers on a phone or hours flowing straight into payroll, buy a dedicated time-tracking product.

Where it goes wrong

• Tracking started with no question to answer, so the data is never used and entry decays.
• A project list that grows without ever being pruned, until picking the right code is the hardest part of the day.
• Retrospective entry at month end, which produces round numbers and no information.
• No approval step, so errors are found by payroll instead of by the manager who could have spotted them.
• Approved periods left editable, which destroys the value of the record the moment anyone touches it.
• Time data used to rank people. It is the fastest way to make every subsequent number untrue.

Employee time tracking is worth exactly as much as the decision it feeds. Pick the question, capture seven columns, approve weekly, lock what is paid — and stop collecting anything nobody reads.