title: "What Held (and What Broke) When 61 Companies Tried a 4-Day Week" slug: "companies-that-tried-a-4-day-work-week" description: "Real results from the UK pilot, Iceland's national trials, Microsoft Japan, and a 141-company global study on the 4-day work week, including who reverted." pillar: "Simplicity & Elimination" secondary_pillar: "Vision-Execution" status: "draft" author: "Marissa Brassfield" keyword_brief: primary_keyword: "4-day work week companies" secondary_keywords:
search_intent: "Informational" people_also_ask:
sources:
url: "https://www.4dayweek.com/uk-pilot-results"
url: "https://autonomy.work/portfolio/uk4dwpilotresults/"
url: "https://autonomy.work/portfolio/icelandsww/"
url: "https://www.cnbc.com/2021/07/06/icelands-shorter-working-week-trial-hailed-as-an-overwhelming-success.html"
url: "https://www.cnbc.com/2019/11/04/microsoft-japan-4-day-work-week-experiment-sees-productivity-jump-40percent.html"
url: "https://www.nzherald.co.nz/business/economy/employment/its-official-four-day-week-becomes-permanent-at-kiwi-company-perpetual-guardian/MB7NO37T25EZDC7IJCY4F7ZXZY/"
url: "https://www.nature.com/articles/s41562-025-02259-6"
url: "https://www.nature.com/articles/d41586-025-02295-2"
url: "https://www.forbes.com/sites/nicksibilla/2025/07/30/new-study-switching-to-4-day-workweek-reduces-burnout/"
url: "https://krystal.io/blog/post/krystal-returns-to-5-day-working-week"
url: "https://www.efinancialcareers.co.uk/news/bolt-4-day-workweek"
url: "https://techcrunch.com/2025/10/27/zoom-ceo-eric-yuan-says-ai-will-shorten-our-workweek/" internal_links:
created_at: "2026-08-10"
Every Slack channel with more than 20 people in it has had this exact conversation. Someone posts an article about a company testing a four-day week, and half the thread runs the same private math: could my team pull off a Friday off? Would my manager laugh me out of the room? Then the thread goes quiet, because nobody expects their employer to run the experiment for real.
In the UK, 61 companies ran it, from June to December 2022, with about 2,900 employees between them. Of those, 56 kept the schedule once the trial ended, and 18 made it permanent policy on the spot. That is not the outcome most of us were braced for.
TL;DR: Companies that tested a 4-day week, from the UK's 2022 pilot to Microsoft Japan, mostly kept it: burnout dropped and retention improved, while output held steady. A smaller group, including Krystal and Bolt, reverted after coverage gaps hurt customers. The five-day week was a guess that calcified into a rule long before anyone measured what it produced.
I have run a 3.5-day workweek since March 16, 2023, the day I declared it on my own birthday and told myself my calendar would never again answer to pressure that wasn't mine. Two decades of working alongside entrepreneurs, three years of documenting my own schedule in public, and a coaching practice built around the audit I describe below give me a specific angle on this research. I read these pilot results as confirmation of something I already tested on myself, at a steeper cut than any of these companies attempted.
The 2022 UK pilot found that cutting one full workday barely moved revenue and moved almost everything else. Across the pilot, 61 companies and roughly 2,900 employees ran the 100-80-100 model (100% of pay, 80% of the hours, a commitment to hold 100% of output) for six months under 4 Day Week Global's tracking, with research support from Cambridge, Boston College, and the think tank Autonomy.
The numbers: 71% of employees reported less burnout and 39% reported less stress, according to Autonomy's full results report. Staff turnover across participating companies fell 57% during the trial period, while company revenue held roughly steady across the same six months.
Nobody ran this pilot to prove a point about wellness. It ran because 61 different management teams wanted to know whether the fifth day was load-bearing. For most of them, it wasn't.
Iceland tested this at a national scale, not a company-by-company one, and the reduction held there too. Reykjavik's city council and the Icelandic national government ran two separate trials between 2014 and 2021, moving more than 2,500 workers from 40-hour weeks to 35- or 36-hour weeks with no cut in pay, across public-sector jobs including hospitals and schools.
Researchers from Autonomy and the Icelandic research group Alda found that productivity held steady or improved across the participating workplaces, and that men in heterosexual relationships took on more household labor once their hours dropped. Service quality did not decline. CNBC covered the trial as researchers called it an "overwhelming success."
The trial's real legacy is what came after it. Icelandic unions used the results to renegotiate contracts nationwide, and roughly 86% of the country's workforce now works reduced hours or has the contractual right to request them. A pilot became a labor standard.
Two of the earliest, most-cited experiments moved the same lever from different directions. Microsoft cut hours company-wide for a month, and a small New Zealand trust company made the cut permanent after a single trial. Microsoft Japan closed its offices every Friday in August 2019 for its Work-Life Choice Challenge, giving roughly 2,300 employees three-day weekends for the month while capping meetings at 30 minutes. Labor productivity rose almost 40% compared with the same month a year earlier, and the company logged 23% lower electricity use, per CNBC's reporting on the trial.
Perpetual Guardian, the New Zealand trust company where founder Andrew Barnes first coined the 100-80-100 formula, ran a two-month trial with 240 employees in 2018. Self-reported work-life balance rose from 54% to 78%, and job stress dropped over the same period. Team engagement scores climbed too, enough that the board made the four-day week permanent company policy rather than an extended experiment.
I coach the same instinct these companies acted on, in individuals instead of HR departments: leaders who assume more hours logged means more value produced, until someone hands them evidence otherwise. That assumption is what most productivity and time-management systems are built to protect. They optimize you to move faster inside a container no one has questioned in decades.
Not every trial held, and the honest version of this story includes the reversions. Krystal, a UK web-hosting company, ran a six-month trial starting in mid-2023 and ended it that September. CEO Simon Blackler said only about half the helpdesk was staffed on Mondays, their busiest day, and the team spent the rest of the week catching up before the pattern repeated. Output didn't rise the 20% needed to offset the lost day, so Krystal added five support staff and reverted, while keeping a shorter daily schedule as a smaller concession.
Bolt, the payments company, ran a four-day week for years starting in 2021 before reversing course and moving employees back to five days. Neither company reported a productivity collapse severe enough to explain the reversal on its own. Both reported the same structural gap: a four-day week redistributes the same five days of demand across four days of coverage, and customer-facing teams feel that seam first.
That is the pattern across nearly every company that reverted. They shortened the week without first shortening the work. The day disappeared. The obligations didn't.
The largest controlled study to date backs the companies that kept it, and it explains why the reverting companies were the exception rather than the preview. Published in July 2025 in Nature Human Behaviour, the study tracked 2,896 employees across 141 companies in Australia, Canada, Ireland, New Zealand, the UK, and the US over six months, with 12 non-participating companies held as a control group. Every company spent roughly eight weeks restructuring workflows before cutting hours, rather than closing the office one extra day and leaving everything upstream untouched.
Employees in the trial reported lower burnout and higher job satisfaction than the control group, along with better mental and physical health. Nature's own news coverage of the study and Forbes's summary both flagged the same detail: about 90% of participating companies kept the schedule permanently after the trial closed, a retention rate that tracks closely with the UK pilot three years earlier.
Zoom's CEO Eric Yuan told reporters in late 2025 that he expects the standard workweek to keep shrinking as AI absorbs more routine work, predicting a three- or four-day week within the next several years. He is describing a future version of the same experiment 141 companies already ran.
A four-day week and my 3.5-day week solve for the same discomfort with two different tools, and the difference explains why some companies keep the shorter week and others quietly give it back. The companies that kept theirs, from the UK pilot to Microsoft Japan, spent weeks before the trial cutting recurring meetings and automating reports, before anyone touched the calendar's day count. The companies that reverted skipped that step and closed the office an extra day, then found the work still waiting on Monday.
I ran that same audit on my own calendar in March 2023, and took it further. I audited which recurring commitments deserved to exist at all, then built the schedule that survived the cut around what remained. The math behind that audit is public, because I would rather show my work than ask anyone to take a shorter week on faith.
None of this required an HR pilot or a company-wide vote. Most coverage of this research treats the four-day week as something only an employer can grant you. The container itself is younger and shakier than it looks, and the standard 40-hour week traces back to a 1938 payroll law, written decades before anyone studied how humans do their best work. Companies willing to question one outdated ritual tend to question the next one too, and the pattern across both is the same. The ritual survives on inertia long after the reason for it disappeared.
Most versions use the 100-80-100 model: employees keep 100% of their pay for working roughly 80% of their previous hours, in exchange for maintaining full output. The specific day off varies by company; Friday is most common.
In the pilots covered here, no. The UK trial, Iceland's national trials, Microsoft Japan, and Perpetual Guardian all held pay steady while cutting hours. Reduced-pay versions exist but are a different, less-studied model.
Perpetual Guardian made it permanent policy after its 2018 trial. In 2023, 18 of the 61 UK pilot companies confirmed permanent adoption, and roughly 90% of the 141 companies in the 2025 Nature Human Behaviour study kept the schedule after six months.
Microsoft Japan measured a 40% productivity jump during its 2019 trial. The UK pilot held revenue roughly steady while cutting hours 20%. The Nature Human Behaviour study found improved well-being with no output collapse across 141 companies. None of the major trials found productivity dropping in proportion to the lost hours.
Krystal and Bolt both reverted after coverage gaps on customer-facing days. In both cases, the company shortened the calendar without first reducing the underlying workload, so the missing day showed up as a staffing crunch rather than a productivity win.
Close, but they differ in degree. A 4-day week typically compresses roughly 32 hours into four days. My 3.5-day week runs 18 to 24 hours across three full days and a half day, built from eliminating the majority of recurring commitments first instead of compressing the same workload into fewer days.
Look at the companies that kept it versus the ones that reverted. The keepers restructured the work before cutting the day. Start there: audit what's recurring on the calendar, cut what doesn't earn its slot, and only then test how few days the remaining work needs.
The UK's 61 companies, 2,896 more in a six-country study, a national trial in Iceland, and a tech company in Tokyo all ran versions of the same experiment, and most of them landed in the same place. The fifth day had survived on habit far longer than it had survived on evidence, and questioning it turned out to be the entire experiment.
Cutting a day off the calendar without first cutting the obligations that filled it is a scheduling trick that borrows against Monday instead of erasing the work. The companies that kept their shorter week did the harder thing first.
What's on your calendar that would survive the same audit, if you ran it before you asked for the day off?
Ready to run your own audit? The Simplicity Protocol sends one elimination strategy a week, the kind you can act on in 20 minutes, for ambitious professionals building a week that doesn't require an employer's permission to work. Subscribe and get the next one.
The world keeps accelerating. The Simplicity Protocol helps ambitious professionals do less to achieve more through weekly elimination strategies you can implement in 20 minutes or less.
Member discussion