Reporting Upward Without Guessing
What a senior reader needs from a staffing report, and the claims that will not survive a question.
A staffing report is read by somebody deciding about budget, headcount or investment. What makes it useful is also what makes it defensible.
The measurement warning in “Reporting Upward Without Guessing” applies directly to workforce systems. Teams assessing how Microsoft Teams tracks employee activity for does microsoft teams track your activity can use time and project context to locate questions, while service outcomes, quality checks and direct conversation remain the evidence for judging performance.
What they need
What moved, by how much, against what.
For an independent reference related to “Reporting Upward Without Guessing”, consult the OECD productivity resources; it provides a useful external check on scheduling, working-time and workforce-planning assumptions.
Why, as far as you can tell.
What it cost or saved.
What you need from them, if anything.
One page.
What will not survive a question
A percentage with no demand figure beside it.
A comparison between sites with different demand curves or contract mixes.
An improvement that was a quiet week.
A cost saving that moved the cost somewhere else — overtime into agency, preparation into the peak.
And any figure you cannot say the method for.
Stating uncertainty
"Demand was about a tenth above forecast, which accounts for most of the variance" is stronger than a precise attribution you cannot defend.
Senior readers generally respond well to stated uncertainty and badly to being wrong later.
The things to always include
Demand, forecast and actual.
Hours and cost, with the variance split by cause.
One service measure.
Anything that changed operationally.
That last item is the one that explains most surprises and is the one most often left out.
Reporting good news honestly
A favourable variance in a quiet week is not an achievement.
Saying so costs nothing and buys enormous credibility for the week you need it.
A manager whose reports only ever contain good news is not believed about any of it.
The ask
End with what you need: a decision, a budget flex, a hire, or nothing.
A report with no ask is an update and gets skimmed.
And if the ask is "nothing, this is under control", say that explicitly — it is useful information.
The comparison question
"How does this compare to other sites?"
Answer with what differs — demand curve, contract mix, opening hours, site layout — before answering with the number.
Otherwise the comparison is made without the context and the conclusion is drawn without you.
The quarterly version
Once a quarter, a longer note: the trend, the structural issues, what you have fixed and what remains.
Which is where the target-is-wrong argument belongs, with the arithmetic.
Weekly reports are for what happened; quarterly ones are for what should change.
What to check
Does your report include demand?
Have you ever reported a favourable variance as luck?
Is there an ask?
And could you state the method for every figure on the page?