Sickness, and the Rate Rather Than the Person
Sickness belongs in a rate as a population figure, never as a prediction about anybody. How to compute it, and why the framing decides whether the number survives.
Sickness enters a labour rate as a population rate applied to a population, in exactly the way an insurance premium does. It is not a prediction about any individual, it does not belong in anybody's personal file, and the moment it is presented per person the figure stops being usable for costing and becomes something else.
The calculation in “Sickness, and the Rate Rather Than the Person” depends on reliable source records as well as a documented accounting rule. For organisations researching employee monitoring software with screenshots, review the platform details here can supply consistent time and project context, while the allocation basis and financial interpretation remain decisions for the organisation and its advisers.
That framing is not sensitivity. It is what keeps the number accurate.
Additional context for “Sickness, and the Rate Rather Than the Person” is available from Sage business resources. Current legal, payroll, tax and accounting requirements should be confirmed for the relevant jurisdiction.
The computation
Paid sickness hours in the last full year, divided by paid hours in the same year, for the same population. One division.
Paid sickness hours in the last full year, divided by paid hours in the same year, for the same population. One division, and the result is a rate rather than a statement about anybody.
Applied forward to the coming year's hours, it gives the sickness layer. It is an average and it is stated as one.
Which population
The team or the grade, not the individual, and not the whole organisation if the teams differ materially. A warehouse and an office have different rates for reasons that are structural.
The unit should be large enough that no individual's year dominates it — roughly twenty people or more. Below that, use the level above.
Why not per person
Because at small numbers the figure is dominated by one or two events, because it has no predictive value for next year, and because a per-person sickness figure inside a costing model will be used for something it was not built for.
A rate computed on twelve people where one had surgery is a statement about that surgery.
Occupational versus statutory pay
Where the employer pays above the statutory minimum, the cost layer is larger and it is the actual paid amount that belongs in the build-up, not the statutory floor.
Schemes often differ by length of service, which means the layer moves as the workforce ages in post.
The cost of cover
Separate from the cost of the absence. Where absence is covered by overtime or agency, that cost belongs on whichever job the cover served, not in the sickness layer.
Counting both the paid absence and the cover in the rate double-counts the same gap.
Long-term and short-term
Worth splitting, because they behave differently. Short-term absence is frequent, small and reasonably stable period to period. Long-term is rare, large, and arrives as a step.
A single rate that blends them is stable in a year without a long case and visibly wrong in a year with one.
Handling the long case
Where a long absence is material against the team's total hours, the honest treatment is to name it as an exceptional item and to state the rate both with and without.
Hiding it inside an average means next year's rate inherits an event that has ended.
Seasonality
Real and usually modest: winter is higher in most operations. For an annual rate it does not matter; for a quarterly one it does, and a quarterly rate built on a winter quarter will be high all year.
Annual computation, annual application, is the simpler answer.
What the figure usually is
In most office-based employments, two to four per cent of paid hours. In physical and shift work, frequently higher. Published national averages exist and describe other people's workforces.
Your own figure is a query against your own payroll, and it is the only one that belongs in your rate.
How to present it
"Sickness: 3.1% of paid hours, team average, from payroll for the year to March. A rate, not a statement about any individual."
The second sentence is doing real work. Without it, the first sentence will eventually appear in a conversation it was not built for, and the next year's data will be worse.
Where the rate is used outside costing
It will be. A sickness percentage computed for a rate build-up appears in a management pack within two quarters, usually without its caveats.
Writing the caveat into the figure's own label — "team average, costing basis" — travels with it better than a footnote, and is the only defence available once the number leaves the spreadsheet it was built in.
Absence cover that is not needed
Not every sick day requires cover. Where the work can wait, the cost is the paid absence and nothing more; where it cannot, the cover cost is on top.
Operations that assume full cover overstate the cost of absence substantially, and ones that assume none understate it in exactly the services where it matters most. The split is knowable from the rostering records and almost never looked at.
Recomputing it each year
Annually, with the rest of the build-up, from the same extract. It moves slowly in a stable workforce and quickly after a reorganisation.
A rate that has not been recomputed since before a major change is describing a workforce that no longer exists.