Other shapes
Compressed hours is not a shorter week
Four ten-hour days removes no work at all, which makes it the safest of the arrangements and the one that produces none of the benefits this record is about.
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.
Compressed hours is the arrangement most frequently reported as a four-day week and it is a different thing. The distinction is not pedantic: the two have opposite risk profiles, and conflating them is why some accounts of shorter weeks report no difficulties whatever.
What it is
The weekly total stays where it was. Forty hours become four tens, or thirty-seven and a half become four at nine and a bit. The fifth day is free; the work is not reduced.
Nothing is subtracted. There is no list of removals, because nothing has to stop.
Why it is safe
Every slow failure on this record traces to a procedure that stopped. Compressed hours produces no such procedures, so it produces none of the five problems, none of the eighteen-month lags and no need for a read-back at all.
An organisation can adopt compressed hours, measure output and satisfaction at six months, find both unchanged or improved, and report accurately that there were no hidden costs — because there were none of the kind this site describes. The six-month window that is inadequate for a four-day week is entirely adequate here.
What it costs instead
Different costs, in three places, and they are not small.
The length of the day. Ten hours is not eight hours with two added. The last two hours of a ten-hour day are not the same two hours as the first two, and the effect is uneven: it falls hardest on physical work, on work requiring sustained attention, and on anybody whose obligations at either end of the day are fixed.
Who can take it. A ten-hour day is unavailable to a large number of people for reasons that have nothing to do with willingness — school hours, caring, a second job, a commute that becomes impossible at the new times, a health condition. A four-day week at thirty-two hours is available to all of those people. Compressed hours is offered to everybody and taken by a subset, and the subset is not random.
Coverage on the fifth day. If everybody takes the same day, the organisation is shut and the obligations that cannot stop still cannot. If the days are staggered, cover exists but the team is never whole, and every procedure that assumes two particular people are available needs revisiting — which is a small list of removals arriving by the back door.
Why it still ends up on the same lists
Published accounts of "the four-day week" frequently include organisations on compressed hours, usually without the distinction being drawn, because from outside the two look the same: people are working four days and reporting that they prefer it.
This matters for reading the evidence. A result drawn from a mixed sample is a result about two arrangements with different mechanisms and different failure modes, averaged. The arrangement with no removals contributes no problems and improves the average; the arrangement with thirty-eight contributes problems that mostly have not happened yet at the time of measurement.
A sample that mixes compressed hours with reduced hours and reports at six months is combining an arrangement with no slow failures and one whose slow failures have not yet arrived.
When compressed hours is the right answer
It is the right answer when the thing being sought is time rather than reduction: a free day, no loss of pay, no reduction in what the organisation does.
It is also the right answer when the work genuinely cannot be reduced — where output is tightly coupled to hours, where coverage is heavy, or where the organisation has recently examined its procedures and there is nothing obvious to remove. On this record the thirty-eight existed because nobody had looked for twenty years. An organisation that has looked has a shorter list and less to gain.
The honest comparison
Set against this record, compressed hours offers:
| four-day week | compressed hours | |
|---|---|---|
| hours reduced | yes | no |
| procedures examined | 38 | none |
| procedures resized | 24 | none |
| slow failures | 5 | none of this kind |
| available to everybody | yes | no |
Four of those five rows favour the arrangement with no removals in it. The row that does not is the one carrying everything else: an organisation that never examines its procedures keeps all of them, and on this record a third of them were doing nothing at all.
The exclusion is the main cost
Of the three costs listed above, the second is the one that gets least attention and does the most damage, and it is worth being specific.
A ten-hour day starting at eight and finishing at six is unavailable to anybody doing a school run at either end. It is unavailable to most people with a caring responsibility that has fixed timing. It is hard for anybody with an hour's commute, because the day becomes twelve. And it is unavailable to several categories of person for health reasons that nobody is obliged to disclose.
A thirty-two-hour four-day week is available to all of those people.
So the two arrangements differ not only in what they remove but in who can take them, and the direction is consistent: the arrangement that asks nothing of the organisation asks a great deal of a particular subset of its staff, and the subset is roughly the one that benefits most from a shorter week in the first place.
An organisation offering compressed hours and reporting high take-up and high satisfaction should look at who did not take it before drawing any conclusion, because the people for whom it does not work simply carry on as before and do not appear in the figures at all.