Other shapes
The nine-day fortnight
Half the reduction, half the removals, and one structural advantage over a four-day week: with staggered days off, nothing ever goes uncovered.
The Friday clean-down of the workshopThe removal a compressed week cannot make
week 9 · crept back after 22 weeks
A named person on the phone until sixThe removal staggered cover is designed to avoid
week 7 · crept back after 5 weeks
Visits to the two furthest sitesThe removal a nine-day fortnight makes instead
week 7 · asked back after 12 weeks
Crept back 2Asked back 1
Three removals were written down with the week each of them stopped. Every one of them produced an event inside the 78 weeks watched. Of these, one was asked back and two crept back. Three removals that a different shape of week would have handled differently.
Nine working days in two weeks. The tenth day off, usually on a rolling basis so that different people are away on different days.
It reduces hours by about half as much as a four-day week, requires about half as many removals, and has one property that neither a four-day week nor compressed hours can offer.
The property
If the days off are staggered, the organisation is never closed and nothing is ever uncovered.
On this record three obligations could not stop: phone cover through business hours, deliveries on somebody else's schedule, and statutory returns on fixed dates. All three were absorbed by four people taking on a longer rota, which is a real cost borne by a small number of people and largely absent from the way the transition was described afterwards.
Under a staggered nine-day fortnight none of that is necessary. There is always a full working day with most of the team present. The obligations are met by the ordinary arrangement rather than by a subset of people doing something extra.
What it costs in exchange
Nobody is off at the same time. The simplification of a whole-organisation closed day — no meetings to reschedule, no cover to arrange, no question about who is available — is gone. Every week contains a different pattern of absences.
Procedures that need two people need rethinking. This is the removal list arriving quietly. Anything that assumed a particular pair would be available now has a one-in-five chance of one of them being away, which is manageable and is not nothing. On this record, two entries — the second reviewer and two-person deliveries — would have been affected this way.
The reduction is smaller, so the forcing is weaker. Four hours a week rather than eight means a shorter list of removals, which means fewer procedures examined. The benefit this record attributes to resizing — twenty-four procedures renegotiated — would be roughly halved.
Where the lag sits
Shorter list, same shape. The removals a nine-day fortnight forces are the same kind of removals, and they will have the same uneven distribution of lags: most reporting inside a fortnight, a few arriving a year later.
There is one difference and it is unhelpful. Because the arrangement never closes the organisation, there is no equivalent of the question that worked best on this record — what did you used to do on the day that went? No single day went. Things that lived in slack still stop, because the slack is reduced, but they do not stop in a way that can be attached to a missing day.
The arrangement that is easiest to cover is the one whose removals are hardest to find afterwards, because none of them is attached to an identifiable gap.
That suggests the compensating practice: the list matters more here rather than less, and the removals have to be written down as they are agreed, because there is no retrospective question that recovers them.
Who it suits
The record points at two situations.
Heavy coverage obligations. Where the thing that cannot stop runs through every working day and cannot be absorbed by a subset of people without a real cost to them, the staggered arrangement is the one that does not create a small group carrying the arrangement for everybody else.
A thin list of candidate removals. An organisation that has already examined its procedures, or that is small enough for everybody to know what everything is for, does not have thirty-eight things to remove. Attempting an eight-hour reduction with twelve candidates means removing things that are load-bearing. Four hours against twelve candidates is a different proposition.
The thing to watch
The halving is not reliably a halving. On this record the removals were not interchangeable units of hours; they ranged from a printed rota to a quarterly reconciliation.
An organisation making half as many removals will not automatically make the safe half. The selection is made by the same meeting, with the same filter described elsewhere on this site, which selects for things that are nameable, defensible and visibly owned — and therefore against exactly the removals that are safe to make quietly.
Half the hours does not mean half the risk unless somebody is deliberately choosing which half.
The pattern problem
One practical difficulty is specific to this arrangement and does not appear in the others, and it is the thing most likely to sink it.
With staggered days off, every week has a different shape. Who is here on Thursday depends on the week. Anything recurring — a weekly meeting, a handover, a pairing — has to either accept that somebody is always missing or move around, and moving around is how recurring things stop happening.
On this record's evidence that is a removal mechanism of its own, and an unusually quiet one. A meeting that nobody cancelled but that half the team misses in alternate weeks is a procedure being removed in instalments, with no entry, no decision and no date.
The organisations that make this work appear to do it by fixing the pattern rather than rotating it: the same person is always off on the same day, so the week has a stable shape even though it is never complete. That gives up the fairness of rotation and buys back predictability, which on the argument of this site is the thing that makes removals visible.