In August 2019, Microsoft Japan closed its offices every Friday, gave its 2,300 staff a three-day weekend on full pay, and capped meetings at 30 minutes. Productivity, measured as sales per employee, rose 39.9 percent against the same month the year before. Electricity use fell. Printing fell. More than nine in ten staff said they were positively affected by the changes.
The story usually gets told as a magic trick. Give everyone Friday off, keep their full pay, and somehow more work gets done, not less. The lesson many people pull from it is the convenient one: just work less and you’ll produce more.
I’ve wanted that to be true for years, which is probably why I don’t quite trust how easily it gets passed around. The real story is less tidy than that. Microsoft Japan’s own write-up of the trial puts as much weight on the meeting cap as on the missing day.
On the surface, it reads like proof that the day off was free.
What actually did the work
The detail many retellings skip is the meetings. It could be argued that the thing that likely moved the number wasn’t the missing Friday on its own. It was the 30-minute cap and the push toward fewer, shorter meetings on the four days that remained. Microsoft didn’t just remove a day and hope. It removed a day and then forced the rest of the week to get tighter.
Isn’t it possible that the free Friday was the reward and the constraint was the mechanism?
That distinction matters, because it’s the part that doesn’t transfer just by wishing. You don’t get the result by simply working less. You get it by being made to work inside a smaller box, so the low-value hours have nowhere left to hide.
I have all the freedom and none of the limit
This is where my own week comes in, and where the story stops being flattering to me. I work for myself, essentially. Nobody schedules my Fridays. If a four-day week is so obviously better, I’m the person least likely to be stopped from taking one, and I don’t.
A while back I tried a soft version of it: wind the work down on Friday afternoons, call a hard stop, claw a bit of the week back. It held for a few weeks. Then “just one more thing” turned the freed afternoon back into an ordinary working block, and the experiment quietly ended without my ever deciding to end it. It went the same way my daily hard stop usually goes.
The freelance version of the problem is that nothing in my day draws a line for me. No office empties out around me, no colleagues stand up and reach for their coats, no commute marks the end. Work just fills whatever space I leave it. The flexibility that’s supposed to be the prize is also the thing that quietly removes every natural place to stop.
Why a limit beats good intentions
Two things seem true to me at once. The first is that a hard limit does what willpower keeps promising to do and doesn’t. Microsoft didn’t ask its people to be more disciplined. It took the day away and shrank the meetings. I’ve come to think the limit is the system, and the system is what actually changes behavior, not the resolve to use the freedom well.
The second is that we probably lose less than we fear. On a good day I can manage maybe three hours of genuinely hard writing; after that the screen is still on, but the real work is finished and the rest is tidying. If a fair amount of the week is slack like that, cutting it doesn’t cost the output it looks like it should.
It should be noted that it wasn’t only a one-summer fluke that only works for a press release. When the UK ran the biggest four-day-week trial to date, with 61 companies, many had kept the policy a year later. Juliet Schor, the sociologist at Boston College who authored the report, said the overall results “have held and in some cases have even continued to improve.”
But the same trial came with a warning that fits my own failure neatly. The clearest benefits seemed to come when the day off was properly protected. Where policies were uneven, conditional, or left people partly on call, the gains were weaker and some staff reported resentment rather than relief.
The lesson here? A limit you might lose isn’t really a limit. It’s just another target. And a limit you set for yourself, with no one holding it but you, is the easiest one of all to quietly hand back.
So here’s the question I can’t shake. If the productive version of the four-day week depends on someone imposing the constraint from the outside, what does it say about us that we need an employer to do it at all? Is this really a story about willpower, or is it a story about the kind of system you happen to be standing inside?
Maybe the uncomfortable answer is that “discipline” was never a personal trait. It was a feature of the environment, and most of us only ever borrowed it from the structures we worked in. Microsoft’s staff didn’t suddenly become better at protecting their time. Their employer protected it for them.
Which leaves the harder question for everyone who works without that scaffolding: if no one is going to take the day away from you, what would you actually have to give up to take it yourself?
