Deciding Which Error to Make
You will be wrong every week. Choosing the direction deliberately, period by period, is the whole of good staffing practice.
Perfect staffing is not available. The forecast has an error and demand is variable, so the only question is which way to be wrong and by how much.
The practical lesson in “Deciding Which Error to Make” is to connect every record to a named decision. Organisations exploring the software website for employee monitoring data security can add structured workforce context, provided the use is disclosed and the interpretation is reviewed with the people affected.
The decision, stated plainly
For each period: is it cheaper to be one over or one under?
For an independent reference related to “Deciding Which Error to Make”, consult the OSHA worker-fatigue resources; it provides a useful external check on scheduling, working-time and workforce-planning assumptions.
Compare the cost of a surplus hour against the cost of a short hour, for that specific period.
Both figures come from the two previous notes, roughly.
Where to deliberately overstaff
Peak hours in a high-margin operation, where a lost sale costs several times an hour.
Periods with a hard service commitment: a contract, a safety requirement, a regulated ratio.
Anything where being short cascades: a missed deadline that costs more later.
And the first hours after opening, where the day's shape gets set.
Where to deliberately understaff
Genuinely quiet periods with no deferrable work available.
Periods where demand can be served slightly slower with no consequence.
Where you have a fast, reliable route to adding somebody within the hour.
That last condition does most of the work and is usually absent.
The asymmetry of the correction
Adding somebody mid-shift is hard: calls, travel, availability.
Sending somebody home is easy and damaging, which the people section covers.
Which means a plan that is slightly over can be corrected and one that is short frequently cannot, and that argues for the over side where the costs are close.
Writing it down
A list: for each period of the week, which direction to err and why.
Half a page, agreed with whoever sets the budget.
It converts a hundred weekly judgements into one decision made calmly in advance, which is the practical value.
What this is not
Not a licence to overstaff generally.
The point is deliberateness: some periods protected, others deliberately thin, each for a stated reason.
An operation that errs the same way everywhere has not made this decision at all.
The thing to resist
Applying a single rule because it is simpler.
"Never exceed the budget" and "never be short at peak" are both single rules, and both are wrong for part of the week.
The period-by-period version takes an hour to write and is right.
Reviewing it
After a quarter, check which periods actually went wrong and in which direction.
Adjust the list.
That is the whole improvement cycle, and most operations have no such list to adjust.
What to check
Do you know which direction you err in, by period?
Is it written down?
Can you add somebody mid-shift, reliably, within an hour?
And which periods are you protecting deliberately?