For a decade the four-day week lived in the space between a good idea and a proven one. Advocates had testimonials and a handful of small pilots; skeptics had the reasonable suspicion that any firm willing to try it was already unusual. In 2025 the argument got its most serious data yet, and the honest reading is that it helps people without settling every question.
What the largest trial found
Fan, Schor, Kelly and Gu (2025), publishing in Nature Human Behaviour, ran a six-month trial of a four-day week with no cut in pay across 2,896 employees at 141 organizations in six countries: Australia, Canada, Ireland, New Zealand, the United Kingdom and the United States. Workers reported lower burnout, higher job satisfaction, and better mental and physical health. Twelve control firms that did not change their schedule showed no such improvement. Covering the study, Nature (2025) noted that employees felt just as productive on four days as on five, with no drop in their own sense of output.
The most useful detail is not the headline but the gradient beneath it. The authors found a dose-response relationship at the individual level: the more a given worker's hours actually fell, the larger their gain in wellbeing. That is a stronger form of evidence than a simple before-and-after, because the effect scales with the treatment. It also came with a plausible account of why. Three mechanisms mediated the improvement: better self-reported work ability, fewer sleep problems, and reduced fatigue. Less time on the clock, more capacity to recover.
Read the limits as carefully as the results
The study earns attention precisely because its authors are frank about what it is not. It is quasi-experimental, not a randomized controlled trial. The companies volunteered, the controls were not randomly assigned, the outcomes were self-reported, and most participating firms were small and English-speaking. One of the authors, the economist Juliet Schor, has said as much in public. Self-selection is the obvious worry: organizations that opt into a four-day trial may already have healthier cultures, more slack, or leaders inclined to trust their teams.
A parallel British trial sharpens the point. CNN (2025) reported on a six-month run organized by the 4 Day Week Foundation, nearly a thousand workers across 17 organizations, after which all 17 kept the shorter week. That retention is striking, but the underlying data are thin and descriptive: only four of the 17 firms supplied revenue figures. Michael Sanders of King's College London, quoted in the same coverage, warned that the firms choosing to participate are not a random sample, which limits what the results can prove about the average workplace. The right way to hold these findings is to take the trial and its critique together, not to pick the half you prefer.
Why the signal still matters
Caveats are not a reason to dismiss the result; they are a guide to how far it travels. What the 2025 evidence establishes is a strong, mechanistically sensible signal that reducing hours can lower burnout and lift health without workers feeling less productive. What it does not establish is that any firm can drop a day and expect identical outcomes, or that the effect is causal in the clean sense a randomized trial would show. Those are different claims, and conflating them is how good evidence gets oversold and then unfairly discredited.
The dose-response finding also reframes the policy question. Because the individual gain grew with the individual reduction in hours, the mechanism seems to be time recovered, not the symbolic appeal of a long weekend. That suggests the lever is real workload, not the calendar label. A four-day week that simply compresses five days of work into four would, on this logic, deliver little, because the hours and the fatigue would not actually fall.
There is also a selection effect worth naming in the other direction. If the firms that volunteer already run healthier cultures, the trial may understate as much as overstate. A company with slack to give back a day is one that had slack to begin with, yet a company drowning in demands is precisely the one whose workers stand to gain the most from recovering time, and precisely the one least able to run the experiment cleanly. The evidence is drawn from the easier cases. That is a reason to test locally rather than assume the published averages will reproduce in a harder environment.
The practical read
For a leader weighing a pilot, the 2025 studies offer a defensible middle path between hype and dismissal. Run it as an experiment, not a announcement. Measure burnout, sleep and self-rated work ability, the mechanisms Fan, Schor, Kelly and Gu (2025) identified, alongside the operational metrics your board actually watches. Expect the wellbeing gains to show up most where hours genuinely drop, and treat any firm-wide claim of no productivity loss as a hypothesis to verify in your own numbers rather than a guarantee imported from someone else's.
The four-day week is no longer a matter of pure faith. It is a matter of evidence that points clearly in one direction while honestly marking how much it still cannot see. That is a better position than the debate has occupied in years, and it is a more useful one than either the boosters or the cynics tend to admit.
Sources
- Fan, Schor, Kelly and Gu (2025). Work time reduction via a four-day workweek. A six-month, six-country trial of 2,896 workers found lower burnout and better health with a dose-response pattern, the primary peer-reviewed evidence for the piece. https://www.nature.com/articles/s41562-025-02259-6
- Nature (2025). News coverage of the largest four-day-week trial. Reported that workers felt just as productive on four days, giving accessible framing of the primary study. https://www.nature.com/articles/d41586-025-02295-2
- CNN (2025). UK four-day week trial coverage. All 17 organizations kept the shorter week, but only four supplied revenue data and Michael Sanders flagged self-selection, supplying the trial and its critique together. https://edition.cnn.com/2025/07/03/business/four-day-week-trial-uk-intl