Reading your work statistics

A monthly summary is only useful if you know what each figure is counting. Most of the confusion people have with their own data comes from four or five numbers that sound interchangeable and are not.

The figures and what they mean

Total hours

The sum of everything recorded as worked. Whether this includes unpaid breaks depends on how you entered them, which is the first thing to check when a total looks wrong. If breaks are recorded as unpaid pauses inside the working interval, they are excluded — which is almost always what you want, because it makes the total comparable against a payslip.

Normal hours and extra hours

The split between contractual time and everything beyond it. This is the pair of numbers to look at first, because it is the pair a payslip is built from. A total that matches while the split does not means the hours are right and the money probably is not — the most common shape of a real error.

Earnings

Hours multiplied by the applicable rates, plus any fixed contributions, minus deductions. This is gross and before tax: it is what you should be paid for the work, not what lands in your account. Comparing it against a net figure on a payslip is the single most common false alarm.

Average per working day

Total hours divided by the number of days with a working entry — not by the number of days in the month. A month with 160 hours across 20 days averages 8.0; the same 160 hours across 16 longer days averages 10.0. The average is the fastest way to notice that a month was composed differently from usual even when its total looks normal.

Effective hourly rate

Total earnings divided by total hours. Where premiums and overtime exist, this sits above your base rate, and how far above tells you how much of your income depends on extra hours. For anyone with more than one job it is the number that makes them comparable.

Month versus year

The two views answer different questions and it is worth being deliberate about which you are asking.

Monthly is for verification. It aligns with the payslip cycle, it is short enough that you remember the individual days, and any discrepancy is still fresh enough to resolve. This is the view for the ten minutes after a payslip arrives.

Yearly is for pattern-finding, and the patterns are usually more interesting than people expect. Total overtime across a year is a workload statement. The seasonal shape of your hours tells you when to expect the crunch. Year-on-year effective rate tells you whether your position has actually improved once premiums are accounted for — a base-rate rise offset by less available overtime is a pay cut that no individual payslip reveals.

The three mistakes behind most discrepancies

When your record and your payslip disagree, the cause is usually one of these, and checking them in order takes about two minutes.

1. Different period boundaries. Your record runs calendar months. Payroll very often does not — a cut-off on the 25th, or a week-based cycle, is common. Under a 25th cut-off, work done on the 28th appears on the following payslip. This produces a discrepancy every single month, in both directions, that is not an error at all. Establish your employer's cut-off date once; it explains a large share of apparent mismatches.

2. Gross versus net. Your record computes gross. A payslip shows gross, deductions and net, and the eye goes to the net figure. Compare gross to gross.

3. A rate that changed and was never updated. If your record still holds last year's rate, your earnings figure has been quietly wrong since the increase. This one is insidious because the hours match perfectly and only the money is off — which looks like a payroll error and is not. Whenever an earnings discrepancy is a constant percentage across several months, suspect the rate before anything else.

Investigating a real gap

Once those three are ruled out, work from the aggregate down to the day:

1. Compare hours first, split into normal and extra. Money follows hours; if the hours are wrong the money question is premature.
2. If total hours match but the split does not, the issue is classification — hours you recorded as overtime were paid as normal, or vice versa. Check the daily threshold used.
3. If total hours differ, look for a whole missing or duplicated day rather than a small drift. Divide the difference by your usual shift length; if it comes out close to a whole number, you are looking for that many days.
4. If the difference is small and irregular, it is usually breaks — a scheduled deduction applied where you recorded an actual one, or the reverse.

What good data looks like after a year

The value of a year of clean records is not any single figure. It is that questions which were previously unanswerable become arithmetic: how much of last year was overtime, whether the second job pays what it appears to, how many hours the commute between sites actually costs, whether your effective rate went up or down after the last increase.

None of those can be reconstructed later. They exist only if the data was collected while it was happening — which is the entire argument for the fifteen seconds a day.