Three on Nights Because There Have Always Been Three
How a staffing number survives the reason for it, the archaeology that finds the original decision, and what to do when the number turns out to be right anyway.
Where one unit's night number came from, traced over an afternoon
A temporary arrangement made during a refurbishment had been running for eleven years and cost a post. Nobody was wrong at any point: the person who added the third post left, and the rota kept reproducing itself. This is one employer's own trace, not a statement of what any standard requires.
The most expensive constraint in most rotas is a number nobody chose: three on nights, two on the early, one on the desk. Everybody can quote it, nobody can source it, and the rota has reproduced it every cycle for years.
The rule discussed in “Three on Nights Because There Have Always Been Three” is easier to test when the published schedule can be compared with a consistent record of actual hours. A team reviewing Monitask for employee scheduling software for employee scheduling software can add current attendance and project context while keeping contractual rules and manager decisions explicit.
These numbers are not arbitrary. They were decided once, for a reason, usually by somebody who has left, and the reason has outlived its own circumstances.
As a separate benchmark for “Three on Nights Because There Have Always Been Three”, see Acas working-hours guidance. Keep the source rule, operational interpretation and scheduling decision distinct.
Why the number survives
Because a rota is built from the last rota. Nothing in the process asks where a number came from; the question only arises when somebody cannot fill it, and at that point the pressure is to fill it rather than to examine it.
The number also acquires authority with age. A figure that has been in use for a decade sounds like a standard, and people defend it as one.
The archaeology
Three sources, in order of speed. The written standard, which either states a number or does not. The longest-serving person in the department, who usually remembers the change. And the minutes or correspondence from whenever the pattern last changed.
- Read the written standard: does it state a number or not?
- Ask the longest-serving person in the department when it changed.
- Find the minutes or correspondence from that change.
- Write down what you found, including "nothing".
An afternoon is enough. In most cases the number is sourced or definitively unsourced within two hours.
The pattern the archaeology finds
Over and over: a temporary increase that became permanent. A refurbishment, a vacancy, an incident, a project, a period of unusual demand — something real that justified an extra person, and no date on which it ended.
Nobody made a decision to keep it. The absence of a decision was the decision.
When the number turns out to be right
Frequently it is. The service has grown, the work is heavier, the layout changed, and the number that was added for one reason is now justified by another.
That is a good outcome and it should be recorded as one: the number, the current justification, the date, and the name of whoever confirmed it. A confirmed number is a completely different object from an inherited one.
When it turns out to be wrong
Then there is a post's worth of capacity in the rota, and a decision to make about what to do with it: reduce, redeploy to a thinner shift, or convert into the training time that removes a skills constraint.
The third is usually the best use and the least likely to be chosen, because the other two are visible in a budget and training is not.
The conversation with the team
This is the difficult part and it has to be handled honestly. A team told that one of their night posts is unjustified will hear a threat, and they will hear it correctly if the exercise is driven by a saving.
The way through is to be clear about the purpose before starting: the register exists to find the constraints that are not real, and what happens to the capacity is a separate decision that people will be told about.
Where safety is raised
Often, and usually sincerely. The response is the same as everywhere else in this collection: which assessment, which hazard, which control.
Where there is a documented control requiring the number, the number is sourced and the exercise is over. Where there is not, the concern may still be right, and it should be escalated as a request to write one rather than dismissed.
The check that would have caught it
A date on every number. Any staffing figure in use should carry the date it was set and the date it was last confirmed, and anything over three years old should appear automatically on a review list.
That one field, applied at the time, would have cost nothing and saved eleven years of a post.
The second number, which is usually the opposite
Departments that have carried an unjustified extra post on one shift very often carry a shortfall on another, for the same reason in reverse: a reduction made during a quiet period that nobody reversed.
Looking for both at once makes the exercise easier to run and much easier to explain. It is a review of every number in the pattern, not a search for savings, and the finding in most places is one of each.
What to write once it is settled
The number, its source or the explicit absence of one, its current justification, and the review date. Four fields, one line per shift type.
It is the shortest section of any constraint register and the one that accounts for the largest sums. Most departments have exactly one of these, and it is usually worth the whole exercise by itself.