The four-day workweek moved from conference talk to controlled experiment, and the experiments now have results. The largest, the U.K. pilot coordinated by researchers at Cambridge, Boston College, and Autonomy with dozens of companies and thousands of workers, documented stable or improved revenue with reduced turnover and burnout, and the majority of participants stayed on the schedule after the trial. U.S. cases exist too, from local-government pilots in Virginia and elsewhere to private firms, and the pattern across them is consistent enough to say what the model is, and is not.
What do the trials actually document?
Three findings recur. Output held: measured revenue per employee and self-reported productivity stayed flat or rose, consistent with the productivity literature's older finding that hours above roughly forty add little output. Well-being moved: burnout scales, stress measures, and turnover all improved, with documented declines in sick days that alone offset part of the cost. And the failure mode was specific: firms with fixed staffing requirements per hour, coverage-dependent services, shift manufacturing, customer-support queues, could not compress hours without hiring, and those pilots stalled, a documented selection effect that means the results describe knowledge-work settings, not the economy as a whole.
How do companies get the same work done in fewer days?
The documented implementations share mechanics. Meeting surgery: cutting standing meetings, shortening defaults, batching the calendar, since meetings are where the recoverable time hides. Focus protection: uninterrupted blocks, async communication replacing status synchrony. Output-based management: setting deliverables and measuring outcomes, which the pilot researchers note is the real work of the transition, and the reason some firms describe the four-day week as a productivity intervention wearing a scheduling costume. Firms that compressed hours without changing work design, the documented failures, saw stress rise as workers crammed five days into four, the model's honest risk.
Is it 32 paid hours or 40 compressed?
The distinction is load-bearing. The studied model is 100-80-100, one hundred percent pay, eighty percent time, one hundred percent output, genuinely reduced hours. The compressed alternative, forty hours in four ten-hour days, is a schedule change, not a workweek reduction, and its documented effects are mixed, liked by some workers for the extra day off, punishing for others in fatigue and childcare arithmetic. Legislative four-day-week bills, introduced in Congress and several states as hours-threshold changes tied to overtime, are really about something else: lowering the overtime threshold at 32 hours, a labor-protection proposal that borrows the brand.
Why has adoption stayed niche?
The documented obstacles are managerial and competitive. Coverage problems, described above, exclude swaths of the economy. Coordination problems: a supplier on four days and a client on five generate friction, which is why the model spreads in clusters, one firm's trial makes its peers' trials cheaper. And trust problems: management-by-presence cultures cannot verify output reductions are real without learning to measure output, a capability gap that is the true rate-limiter, the case studies argue, not worker resistance, which the pilots found minimal once schedules landed. Labor-market power matters too: the pilots clustered in tight markets for professional talent, where the four-day week functions as a recruiting edge, exactly as remote work did before it normalized.
What should a worker or employer take from the record?
Workers: the evidence supports asking, since the documented pilots show output held and well-being rose, and the firms that tried it mostly kept it. Employers: the transition cost is real, weeks of process redesign, and the payoff is documented in retention, not just recruitment, with turnover the expensive thing the schedule quietly fixes. The four-day week is neither a utopia nor a fad on this record; it is a bundle of management reforms that some jobs can run and others cannot, and the honest question for any workplace is not whether shorter weeks work, but whether the work has been designed well enough to try.
For more context, read What the WARN Act Actually Owes Workers When a Plant Closes.
For more context, read childcare costs data.
For more context, read Your Remote Job Has a Tax Map Problem.