The 4-day work week: what the UK, Iceland, and Portugal trials actually show

60-Second Summary

Most published '4-day week trials' use the 100-80-100 model: 100% pay, 80% time, 100% output. They are not '32-hour-week-instead-of-40 with proportional pay cut'.
Iceland 2015–2019 (the most rigorous): 2,500+ public-sector workers reduced to 35–36 hours. Wellbeing improved across all measures; productivity stable or improved in most workplaces. Now 51% of Icelandic working population on shorter weeks.
UK pilot 2022 (Autonomy / 4 Day Week Global / Cambridge / Oxford / Boston College): 61 companies, ~2,900 employees. After 6 months: 56 of 61 companies continued; revenue +1.4% vs same-period prior year. Burnout, anxiety, fatigue all down. Methodology criticised for no proper control group.
Portugal pilot 2023: 41 companies, government-supported. Burnout dropped from 16% to 8%; sleep quality improved; revenue stable. Smaller sample, similar methodology issues.
Belgium 2022: legal right to request a compressed 4-day week (same hours, 4 days) — different policy entirely; uptake low (~1%).
Honest read: the wellbeing evidence is strong and replicated. The productivity claims are credible but methodologically weak (no real control). Works best for knowledge work, focused teams, with serious workflow redesign. Doesn't translate to shift work, hospitality, retail, healthcare without major redesign.

The 4-day work week is one of the most-discussed and most-misunderstood workplace experiments of the 2020s. This guide cuts through the press releases to the actual evidence base, the methodological caveats, and what to do if you want to run a pilot in your own org.

Three different 4-day week models

Model	Hours	Pay	Where it's been tested
100-80-100 (reduced hours)	32 (vs typical 40)	100% maintained	UK 2022, Portugal 2023, US pilots
Compressed (4×10)	40 hours over 4 days	100% maintained	Belgium 2022 right-to-request
Reduced + reduced pay	32 hours	Reduced proportionally	Rare in formal pilots; common informally

When people talk about 'the 4-day week pilots', they almost always mean the first model. The Belgian model is structurally different and shouldn't be conflated with it.

Iceland 2015–2019 (the rigorous one)

The most rigorous and longest-running trial. Reykjavík City Council and the Icelandic government reduced working hours for 2,500+ public-sector workers (1% of the Icelandic workforce) from 40 to 35–36 hours per week with no pay reduction, across diverse roles (offices, preschools, social services, hospitals).

What Iceland actually found

1
Wellbeing
Improvements across every measured dimension — work-life balance, stress, burnout, sleep, family life. Effects sustained over multi-year follow-up.
2
Productivity
Stable or improved in most workplaces. Critical that the trial was not just shorter hours but accompanied by genuine workflow redesign (shorter meetings, fewer interruptions, deferred non-core tasks).
3
Long-term impact
Trade unions used the trial results to negotiate shorter hours into collective agreements. By 2022, ~51% of the Icelandic working population had won the right to shorter weeks; ~86% have either won shorter hours or have the right to renegotiate.
4
Honest limitation
Public sector dominant; private-sector translation requires separate evidence.

UK 2022 pilot

The most-cited international trial. 61 companies, ~2,900 employees, 6 months (June–Dec 2022). Coordinated by 4 Day Week Global, Autonomy, with researchers at Cambridge, Oxford, and Boston College.

UK pilot — what the published reports say

1
Continuation
56 of 61 companies continued the 4-day week after the pilot; 18 made it permanent immediately.
2
Business outcomes
Revenue +1.4% during pilot vs same period prior year (weighted average). Resignations dropped 57% vs comparison period.
3
Employee wellbeing
Burnout −71%, anxiety down, fatigue down, sleep quality up. Job satisfaction up across nearly all measures.
4
Methodology caveats
Self-selected participating companies (likely positively biased). No randomised control. Self-reported wellbeing. Revenue comparison was company-specific, not market-adjusted. The signs are strong but the design wouldn't pass a tier-1 economics journal.

Read what's there, not the press release

The wellbeing evidence is robust enough to act on. The productivity claims are directionally credible but should be read with humility about methodology. 'No drop in productivity' is the honest finding; '+1.4% revenue caused by 4-day week' is overclaiming.

Portugal, Spain, Belgium

Portugal 2023: 41 companies, government-supported pilot. Published results June 2024. Burnout dropped from 16% to 8% (8pp). Sleep, mental health, perceived productivity all improved. Revenue stable. Smaller sample, similar methodology.
Spain 2023: 200+ companies signed up for government-supported pilot (EUR 10M earmarked). Mixed implementation; not yet rigorously published at scale.
Belgium 2022: legal right to request a compressed schedule (4×10) with employer's reasonable consent. Different model entirely — same hours, just compressed. Uptake reported ~1%; most workers prefer the genuine reduction.
Iceland (continuing): post-trial, multi-year follow-up suggests gains persist over 5+ years.

What seems to work and what doesn't

Pattern across pilots

Where it works

Knowledge work with controllable schedules
Companies that redesign workflows, not just cut a day
Teams with high-quality meeting hygiene
Cultures where focused work is valued over visible busyness
Public sector and professional services strongest results

Where it struggles

Shift work, healthcare, hospitality, retail (without doubling staff)
Companies that cut a day but keep all meetings
Client-facing teams with global coverage requirements
Sales teams on activity-based metrics
Industries with tight regulatory cycles (e.g. period-end finance, audit)

Implementation playbook

Pilot scope: one team or BU of 50–500 people, 6 months minimum. Set the comparison group up front (matched team or pre-period data).
Workflow redesign: weeks 1–4 before the pilot — kill 30–50% of recurring meetings, set focus-time blocks, redesign approval chains. Without this, the 4-day week is just 5 days of work in 4.
Pay maintained: 100-80-100. Cutting pay defeats the experiment and gives noisy results.
Measurement: define 3 business outcomes (revenue, throughput, quality, customer outcome) and 3 wellbeing outcomes (burnout, sleep, work-family conflict). Measure pre, mid, post.
Honest report-out: publish results internally including what didn't work. Companies that suppress mixed results lose credibility for future pilots.
Decision after pilot: continue, modify, or stop. Most successful pilots evolve into a 9-day fortnight or rotating model rather than strict Friday-off.

FAQ

Frequently asked questions

Do we need to give Fridays specifically?

No. Most pilots find rotating off-days (so the business covers Monday–Friday) work as well as fixed Fridays. For client-facing teams, rotating is usually required.

What about meetings with customers and partners on the off-day?

Largest single source of pilot failure. Either coverage rotation, or a designated 'emergency on-call' from the off-team, or accept the friction. Companies that don't decide this in advance burn out the people who end up working anyway.

Does it work for engineering teams?

Yes — engineering is one of the strongest-result populations. Focus work benefits the most from the redesign. Many engineering teams find that 4 focused days produce more shipping than 5 interrupted ones.

What about manager bandwidth?

The most under-discussed issue. Managers with 8+ reports often struggle to compress 1:1s and admin into 4 days. Pilots should either reduce manager spans or formally protect manager admin time.