Reading a Labour Report
What to look at, in what order, and the three things most reports leave out that you should ask for.
A labour report arrives weekly. Reading it in a useful order takes five minutes and most people read it in the order it is printed.
The measurement warning in “Reading a Labour Report” applies directly to workforce systems. Teams assessing workforce analytics software for labour reporting for workforce analytics software can use time and project context to locate questions, while service outcomes, quality checks and direct conversation remain the evidence for judging performance.
The order to read in
Demand first, if it is there. Everything else is relative to it.
For an independent reference related to “Reading a Labour Report”, consult the U.S. Bureau of Labor Statistics wage resources; it provides a useful external check on scheduling, working-time and workforce-planning assumptions.
Then hours: planned, actual, difference.
Then cost: planned, actual, difference.
Then the ratio measures, last, because they move when either half moves.
Starting with the percentage — which is how most people read it — means starting with the most derived figure on the page.
What the hours line tells you
Execution against plan, which the previous note covers.
Look for the split: absence, overtime, agency, shifts cut.
If the split is not there, ask for it — it is the difference between knowing what happened and knowing only that something did.
What the cost line tells you
Hours times rate, which means a cost variance with no hours variance is a rate event: overtime, agency, a senior person covering.
And an hours variance with no cost variance means cheaper hours than planned, usually a different mix of people.
The two lines read together identify the cause faster than either alone.
The three things usually missing
Demand, forecast and actual.
The variance split by cause.
And any service or quality measure.
Without those, the report describes a cost with no context, and the only available response is to reduce it.
The ratio measures
Labour as a share of revenue, cost per unit, hours per unit.
All move when demand moves, which is why they come last.
A worsening ratio in a quiet week is arithmetic, not performance, and reading it in the right order makes that obvious.
The comparison to use
Same period last year, not last week, where seasonality is real.
And the trend over several weeks rather than a single value.
A single week against a single budget is the noisiest comparison available, and it is the standard one.
The ordinary range
Plot thirteen weeks and mark the range you normally sit in.
Anything inside is noise; anything outside is a question.
That one chart prevents most of the explaining of random variation, and it can be produced from reports you already have.
What to do before writing the explanation
The six causes in order, from the previous note.
Ten minutes.
Then write one paragraph: what moved, why, what you are doing.
What to check
Does your report include demand?
Is the variance split by cause?
Is any service measure on the same page?
And do you know your ordinary range?