4 Day Work Week: What the Trials Actually Measured
4 Day Work Week: What the Trials Actually Measured
The best evidence on the four-day work week says it improves how workers feel: lower burnout, better sleep, less fatigue. In the firms that reported revenue (23 out of 61 in the largest UK pilot), output mostly held steady. But no trial was randomized, every participating company volunteered, and most outcomes were self-reported. The version of "four-day week" matters too: 32 hours at full pay is a different intervention than 40 hours crammed into four 10-hour days.
Key takeaways:
- The two biggest trials (UK 2022 and Fan et al. 2025) measured well-being, not productivity. Revenue data came from fewer than half the UK firms.
- Every major trial used self-selected companies. The authors of Fan et al. say their findings likely overestimate the true effect.
- A "4 day work week" can mean 32 hours at full pay (the 100-80-100 model) or a compressed 4x10 schedule. The health and focus implications are different.
- The firms that succeeded did not just delete Friday. They cut meetings, moved communication async, and protected focus time first.
The Big Trials at a Glance
| Trial | Who | Hours Change | What Was Measured | Headline Result | Main Limitation |
|---|---|---|---|---|---|
| UK pilot (2022) | 61 UK organizations, ~2,900 workers | 40 to ~32 hrs (20% cut) | Self-reported burnout, stress, sick days; revenue for 23 firms | 71% lower burnout; revenue +1.4% avg (23 firms) | No control group; firms self-selected through advocacy org |
| Fan et al. (2025) | 141 organizations, 2,896 workers, 6 countries | 39.21 to 34.01 hrs avg (~13% cut) | Burnout, job satisfaction, mental/physical health | Burnout dropped from 2.83 to 2.38 (1-5 scale) | Not randomized; 12 control companies only; self-selected sample |
| Iceland (2015-2019) | 2,500+ public-sector workers | 40 to 35-36 hrs | Well-being, service provision, work-life balance | Maintained or improved service; 86% of workforce later moved to shorter hours or gained the right to them | Five shorter days, not four; public sector only |
| Microsoft Japan (2019) | One company, one month (August) | 5 Fridays off | Sales per employee vs Aug 2018 | +39.9% sales per employee | Company-reported; one month; concurrent 30-min meeting cap |
| Bambra (2008) | 40 studies of compressed weeks | Same total hours, fewer days | Self-reported health, work-life balance | No detrimental effect on self-reported health in 5 controlled studies; work-life balance improved | Mostly self-reported; low methodological quality across studies |
What Is a 4 Day Work Week? Two Models, Not One
The phrase "4 day work week" describes two structurally different schedules.
The 100-80-100 model means 100% pay, 80% time, 100% expected output. Workers drop to about 32 hours across four days. This is what the UK pilot and Fan et al. tested.
The compressed 4x10 schedule keeps total hours at 40 but packs them into four 10-hour days. Tokyo's Metropolitan Government adopted this model in April 2025, allowing staff to take three days off per week under a flexible system with total monthly hours of 155. Same hours, rearranged.
In the US, the Thirty-Two-Hour Workweek Act (S. 3947) was introduced in March 2024 by Senators Sanders and Butler. It would lower the FLSA overtime threshold from 40 to 32 hours in phases. It has been introduced. It is not law.
The health trade-offs diverge between models. Bambra's 2008 review of compressed schedules found that work-life balance improved, but the studies were low quality and outcomes mostly self-reported. A 10-hour day can bring fatigue that a 32-hour week avoids.
What Did the UK Four-Day Week Trial Actually Find?
From June to December 2022, 61 UK organizations and about 2,900 employees participated in a six-month pilot studied by researchers at the University of Cambridge and Boston College with the think tank Autonomy. Workers took a 20% reduction in hours with no reduction in pay.
The results, by the numbers:
- 71% of employees self-reported lower burnout
- 39% said they were less stressed compared to the start of the trial
- 65% reduction in sick days vs. the same period the prior year
- 57% fall in staff turnover vs. the same period the prior year
- Revenue rose 1.4% on average, but only 23 of 61 organizations provided revenue data. That denominator matters.
- 56 of 61 companies (92%) intended to continue; 18 confirmed the policy as permanent
One year later, the follow-up was strong. Autonomy's "Making It Stick" report found at least 54 of 61 firms (89%) were still operating a four-day week, and at least 31 (51%) had made it permanent.
What this trial did not have: a control group. Every firm entered voluntarily through an advocacy organization. The 92% continuation figure tells you something about these specific firms, not about all firms.
The Biggest Study Yet: 141 Companies in Six Countries
Fan, Schor, Kelly, and Gu (2025), published in Nature Human Behaviour, followed 2,896 employees across 141 organizations in six countries through a six-month trial. Twelve companies served as controls.
Average company hours fell from 39.21 to 34.01 per week, a cut of about 13%, not the full 20% the label implies. Work burnout fell from 2.83 to 2.38 on a 1-to-5 scale. Job satisfaction, mental health, and physical health all improved. The control companies showed no comparable changes.
The mediators: self-reported work ability improved, sleep problems fell, fatigue decreased. The gains were not about more output per hour. They were about less damage per week.
The authors are explicit about the limits. Firms were recruited through 4 Day Week Global, an advocacy organization. The sample is mostly white-collar. The authors themselves write that their findings likely overestimate the true effect.
Iceland's "Four-Day Week" Was Actually a 35-to-36-Hour Week
The Iceland trials, run by Reykjavík City Council and the national government from 2015 to 2019, involved more than 2,500 workers, about 1% of the Icelandic workforce. Hours dropped from 40 to 35 or 36 at the same pay.
For many participants, that meant five shorter days, not four. Calling it a "four-day week" is media shorthand.
Service provision was maintained or improved, and well-being rose. The political impact was large: according to Autonomy and the Association for Sustainable Democracy (Alda), about 86% of Iceland's workforce has since moved to shorter hours or gained the right to negotiate them. A policy success, but one about reducing weekly hours, not about a specific day off.
Does a 4 Day Work Week Increase Productivity?
This is the question everyone asks, and the answer is less clean than the headlines suggest.
Microsoft Japan gave employees five Fridays off in August 2019, capped meetings at 30 minutes, and pushed communication online. Sales per employee rose 39.9% versus August 2018. But this was one company, one month, with concurrent process changes. A case study, not a controlled experiment.
The broader evidence on hours and output is older but stronger. Economist John Pencavel, analyzing data on British munitions workers from World War I, found that below about 49 hours per week, output was roughly proportional to hours. Above that, returns diminished sharply. Cutting from 55 to 40 costs less output than you would expect. Cutting from 40 to 32 is a different question, and Pencavel's data does not answer it.
A 2021 WHO/ILO study estimated that working 55+ hours per week led to 745,000 deaths from stroke and ischemic heart disease in 2016: 35% higher stroke risk, 17% higher heart disease mortality versus 35-to-40-hour weeks. That is the case against long weeks, not proof that shorter weeks improve output.
4 Day Work Week Pros and Cons
| Pros (supported by trial data) | Cons and unknowns |
|---|---|
| Lower self-reported burnout (UK: 71%; Fan et al.: 2.83 to 2.38 on 5-point scale) | All major trials used self-selected firms; no randomization |
| Better sleep and reduced fatigue (Fan et al. mediators) | Most outcomes self-reported, not independently measured |
| 65% fewer sick days in UK pilot | Revenue data from only 23 of 61 UK firms |
| 57% less staff turnover in UK pilot | Compressed 4x10 carries fatigue and childcare trade-offs |
| Improved work-life balance across trials | Hourly and shift workers are underrepresented in samples |
| High retention: 89% of UK firms still on it at one year | Customer-facing and coverage-dependent roles face scheduling gaps |
| Bambra review: compressed weeks improved work-life balance | Bambra review: studies were low quality, mostly self-reported |
What the Trials Changed Before They Cut the Day
The companies that made four-day weeks work did not just cross Friday off the calendar. They redesigned how work happened first.
Microsoft Japan capped meetings at 30 minutes. Fan et al. describe "pre-trial work reorganization" in which companies improved efficiency and collaboration before the six-month clock started. The common thread: fewer meetings, async communication, and protected focus blocks. These are the changes Parkinson's law predicts: work expands to fill the time, so when you cut the time, you cut the filler.
Whether your company runs four days or five, the practical lesson is the same. Your output tracks your focused hours, not your logged hours. Protecting two or three deep work blocks per day from meetings and interruptions is the part of these trials any schedule can copy, even if the headline was about the missing Friday. For more on guarding those blocks, see what context switching actually costs and timeboxing your focus hours.
So, Does the 4 Day Work Week Work?
For well-being in volunteer, mostly white-collar firms: yes, consistently. Burnout drops, sleep improves, fatigue falls, people stay longer.
For output and revenue: probably neutral in the firms that tried it, but the data is thin (23 firms), self-reported, and drawn from companies already optimistic about the idea.
For everyone: unproven. Hourly workers, shift workers, and industries with hard coverage requirements were mostly absent. And the compressed 4x10, which is what many employers actually mean by "four-day week," is a different intervention with different trade-offs. The version matters more than the label.
Frequently Asked Questions
Is a 4 day work week 32 hours?
Usually, but not always. The 100-80-100 model tested in the UK pilot and Fan et al. means about 32 hours at full pay. A compressed 4x10 schedule keeps total hours at 40. Tokyo's public-sector model uses 155 hours per month. Same label, different hours.
Which companies have a 4 day work week?
Microsoft Japan trialed it for one month in August 2019. The Tokyo Metropolitan Government began offering a compressed option in April 2025. Beyond those, the best-documented examples are the 61 UK pilot firms (56 intended to continue) and the 141 volunteer organizations in Fan et al., mostly white-collar and recruited through 4 Day Week Global.
Do you get paid the same on a 4 day work week?
In the 100-80-100 model, yes. Workers take a 20% reduction in hours with no cut in pay. The compressed 4x10 model also preserves full pay since total hours remain at 40. Both the UK pilot and Fan et al. held wages constant.
Is a 4x10 schedule better than five 8-hour days?
Bambra's 2008 review found that compressed weeks improved work-life balance, but the studies were low quality and outcomes were mostly self-reported. A 10-hour day can bring more fatigue, and childcare logistics can be harder. Whether it is "better" depends on your role, your commute, and how well you sustain focus past hour eight.
Did productivity drop in the four-day week trials?
Revenue in the 23 UK firms that reported it rose 1.4% on average. Fan et al. measured well-being, not revenue. Microsoft Japan's 39.9% jump was one company, one month, with concurrent changes. Output mostly held steady where measured, but it was not measured in most trials.
Is the US moving to a 32-hour work week?
The Thirty-Two-Hour Workweek Act (S. 3947) was introduced in the Senate in March 2024. It would phase the FLSA overtime threshold down from 40 to 32 hours. It has not advanced beyond introduction and is not law. No federal policy change is imminent as of 2026.
Four Days Only Work If the Focus Hours Hold
The trials did not just delete a day. They cut meetings, moved communication async, and built protected focus time into the remaining hours. The well-being gains tracked through less fatigue and better work ability, not through some abstract productivity multiplier.
That is the part that transfers to any schedule: your output depends on how many of your hours are genuinely focused, not how many hours you log. On a compressed 4x10, or a standard five-day week, or an actual 32-hour week, the bottleneck is the same.
Roon is built for those focus blocks. Each pouch delivers 80 mg caffeine, 60 mg L-theanine, 25 mg methylliberine (Dynamine), and 5 mg theacrine (TeaCrine) sublingually, with onset in 5 to 10 minutes and a felt window of 6 to 8 hours. The duration comes from methylliberine slowing caffeine's clearance: in a randomized crossover study (n=12), co-administration cut caffeine's oral clearance from 41.9 to 17.1 L/hr and extended its half-life from 7.2 to 15 hours, with no increase in peak concentration. That study used 100 mg methylliberine and 150 mg caffeine, roughly 4x and 1.9x Roon's doses, so the direction applies but the exact numbers do not. The felt window is 6 to 8 hours, not 15.
A longer half-life also means you should stop at least 8 hours before bed. On a 10-hour day starting at 7 AM, that means taking it early. Roon is not a way to stretch the workday. It is a tool for the focus hours you protect, starting from $9.16 per tin on subscription ($0.61 per pouch).
Related from Roon
- Parkinson's law: work expanding to fill the time
- What context switching actually costs
- Timeboxing your focus hours
- Burnout recovery: what actually works
Written by Roon Team
