Written Down, Said Once, or Just Assumed
The three states a rota rule can be in, why the middle one causes most of the trouble, and how to move a rule from one state to another without losing it.
A rule about a rota exists in one of three states: written in a document somebody can produce, said once to somebody who remembers it, or assumed by everybody and said by nobody. The three behave entirely differently when questioned.
The constraint in “Written Down, Said Once, or Just Assumed” becomes useful only when its source, cost and effect on coverage are visible. A team considering how should teams assess task prioritization methods? for task prioritization methods can add timely work and attendance context, with transparent settings, limited access and a clear correction route.
Most of the trouble sits in the middle state. Something said once — at an interview, in a corridor, during an induction — is real to the person who heard it and invisible to everybody else.
For an outside reference relevant to “Written Down, Said Once, or Just Assumed”, consult Cochrane shift-work evidence. Compare the general material with the organisation’s jurisdiction, sector and written agreements.
Written
The easiest to handle, because it can be read. A written rule has a version, usually a date, and sometimes an author, and all three can be checked against what the rota actually does.
Written does not mean binding, and this is the common confusion. A document the organisation wrote for itself is a decision it can revisit; a document somebody else wrote may not be. The distinction is the one that matters, and it is a different question from whether it is written.
Said once
A promise, an assurance or an instruction given in conversation. "You'll never be asked to do nights." "We can keep Thursdays clear for you." "Always have two on when the furnace is running."
These are real in their effects and invisible in the record. The person who heard it plans their life around it; the rota system knows nothing; the manager who gave it has moved on. The first anybody hears of it again is when it is breached.
Assumed
Nobody said it and everybody believes it. Assumptions grow out of repetition: a pattern that happened to hold for two years becomes a rule by the third, and the rule acquires a reason afterwards.
Assumed rules are the cheapest to test and the most awkward to raise, because asking looks like a challenge to whoever is assumed to have decided. Framing matters: the question is what the rule is, not who got it wrong.
Moving a rule between states
The useful direction of travel is from said and assumed towards written. The work is small:
- Ask the person who holds the belief what exactly it is.
- Ask what it is based on, accepting "nothing I can point to" as an answer.
- Write the rule in one sentence, with whatever source exists.
- Decide, as a live question, whether to keep it.
- Date the entry and name who confirmed it.
The decision step is the one people skip. Writing down a rule nobody wants to keep is how organisations end up with documented nonsense.
What this does to individual promises
A promise made to one person should not quietly become a rule for a team, and it should also not evaporate because the manager who made it has left.
Writing it down achieves both: the arrangement becomes a recorded individual arrangement with a review date, which protects the person and stops the rota treating it as a general law. That is better for everyone involved, including the organisation.
Why not everything should be written
A rota does not benefit from a rulebook. Some constraints are better handled as judgement, and formalising them converts a flexible practice into an obligation that is harder to vary than the thing it replaced.
The test is whether somebody would be harmed by the rule being silently dropped. If yes, write it. If no, record it as practice and leave the judgement where it is.
The audit worth running once
Take the current rota, list every constraint it visibly implements, and mark each W, S or A. Most managers can do this in an hour, and most are surprised by the proportions.
An organisation with a third of its constraints in S or A is normal. An organisation that believes it has none is one that has not looked, which is the state nearly all of them are in before the exercise.
What good looks like afterwards
Not a thick document. A page where every line is either sourced or honestly marked as unsourced, and where the unsourced lines have a date and a name against them.
That page survives a manager leaving, which is the only real test of whether a rota constraint set exists at all.
The entry that is always missing
Whatever was agreed with an individual, in writing, and reviewed since. Flexible arrangements are recorded at the point they are granted and then never looked at again, by anybody, in either direction.
That is the single most common gap in these lists, and it is the one that produces the most uncomfortable conversations when it finally surfaces.