Rebuilding the Rate Annually
A half-day exercise that decides every job cost for a year: the eight inputs, where each comes from, and what to do with the result.
Rebuilding a labour rate is a half-day exercise with eight inputs, and it decides every job cost and every quote for the following year. It is also, in most organisations, a figure inherited from a predecessor and adjusted by inflation.
The calculation in “Rebuilding the Rate Annually” depends on reliable source records as well as a documented accounting rule. For organisations researching task switching cost, task switching cost can supply consistent time and project context, while the allocation basis and financial interpretation remain decisions for the organisation and its advisers.
The eight inputs, each with a source and a date, are the whole of it.
Additional context for “Rebuilding the Rate Annually” is available from Investopedia business reference. Current legal, payroll, tax and accounting requirements should be confirmed for the relevant jurisdiction.
The eight inputs
- Wage cost, from payroll.
- Employer contributions, from payroll.
- Pension, from payroll.
- Paid absence hours, from payroll and the calendar.
- Sickness rate, from payroll, last full year.
- Training hours, from the booking function.
- Per-head costs, from the asset register and licence contracts.
- Chargeable hours, from the job system and normal capacity.
Seven queries and one judgement.
The judgement
Normal chargeable hours. Everything else is a reading; this one is a decision about what a typical year looks like, and it should be made deliberately and recorded with its reason.
It is also the input that moves the answer most.
The order
Inputs first, all eight, before any arithmetic. Assembling them while computing invites adjusting one to make the answer look better, and the adjustment will be invisible a year later.
The output
A build-up table from wage to rate, every line with its source and its date, and one sentence naming the divisor.
One page. It should be publishable to anybody who asks what an hour costs, and it should survive their questions.
When to do it
After the pay award and before the budget, which in most organisations means a different month from the one it currently happens in.
Rebuilding before the award means rebuilding twice or carrying a known error for most of the year.
How many rates
One per group whose cost or chargeability differs materially. Usually three to six: by grade, by site where premises differ, by department where the work differs.
Twenty rates is administration. One rate across a mixed workforce is a blend presented as a fact.
Testing it before publishing
Multiply the rate by budgeted chargeable hours and compare to the loaded cost of the department. They should match.
Where they do not, something has been double-counted or omitted, and finding out before publication is considerably better than after.
The second test
Apply it to three jobs from last year whose outcome is known. If a job everybody remembers as comfortable comes out at a loss, something is wrong with the rate or with everybody's memory, and both are worth knowing.
What to do with the change
Say what moved and by how much. "Rate up 4.1%: pay award 3.4%, chargeable hours down 1.2%, per-head costs up on licences."
Three causes, named. A rate that changes without an account of why gets treated as a finance department decision rather than as a measurement.
Keeping the old ones
Every version, with its dates and its build-up. Reconstructing a job cost from two years ago is asked for more often than expected and is impossible without them.
What it replaces
The annual argument about whether the rate is too high, which is otherwise conducted on impressions and settled by whoever is most senior.
With the build-up on the table, the argument becomes a line-by-line discussion of real costs, and it usually ends with agreement and two corrections.
Who should be in the room
The person who owns the costing, somebody from operations who knows the work, and somebody who can speak for payroll data. Three people, half a day.
Built by finance alone the divisor will be wrong; built by operations alone the layers will be. The half-day is also where the two functions discover what the other has been assuming, which is usually the most valuable part.
The inputs that are hardest to get
Training hours and per-head costs, both of which live outside payroll and outside the job system, in functions that have no reason to produce them.
Asking once a year, with a specific format and a date, works. Asking vaguely produces a number somebody estimated in the meeting, and that estimate then sits in the rate for a year with nothing behind it.
What the rebuild usually finds
One layer that has moved materially and nobody flagged — most often licences or pension — and one input that nobody can source any more because the person who produced it has moved on.
Both are arguments for the build-up being a document with named sources rather than a spreadsheet somebody maintains.
The maintenance between rebuilds
None, deliberately. The rate is fixed for the year. What is maintained is the variance record, which feeds next year's rebuild.
Stability is the point. A rate that is adjusted whenever somebody questions it is not a measurement.