Perpetual Guardian Four-Day Week: Full Pay, One Day Less, Same Output?

The room

The psychology: goal-setting theory

Edwin Locke and Gary Latham built the theory over 35 years of laboratory and field studies. Their 2002 review describes four ways a goal works. It directs attention towards what matters for the goal and away from what does not. It energises effort, because harder goals produce more effort than easy ones. It prolongs persistence. And it pushes people to find better strategies. The effects are strongest when people are committed to the goal and get feedback, and weaker on complex tasks where people first need to learn how to do the work.

The Perpetual Guardian trial turned a vague aim, being more productive, into a specific, difficult and public goal: deliver the same output in 30 hours instead of 37.5. The reward was concrete and personal, a day off each week. Staff helped define what productivity meant in their roles, which builds commitment, and the trial’s measures gave feedback. The changes staff reported, such as shorter meetings and fewer distractions, are the strategy mechanism at work: a hard limit forced people to look for waste and remove it.

The same mechanisms explain the trial’s less visible side. If output is fixed and time shrinks, effort per hour has to rise somewhere. Research on flexible working calls this work intensification, and it can happen even when people are happy with the deal, because they want to give something back for a benefit they value. The 100:80:100 idea reads very differently depending on where the extra output comes from: removing waste, or simply working harder in fewer hours.

Why smart people fall for it

Three forces explain why a paid day off changed how people worked, and why the change was not free.

1. 1

   A hard limit forced better strategies

   With 30 hours to deliver 37.5 hours of output, teams had to find and cut waste such as long meetings and interruptions.

2. 2

   A personal reward built commitment

   The day off was visible, valued and conditional on output, so staff had a clear reason to hit the goal.

3. 3

   Gratitude raised effort per hour

   Staff called the day a gift, and some said they were willing to be available for work on it, which is effort the headline numbers do not show.

Did the Perpetual Guardian four-day week really keep output at 100 percent?

The careful answer is that output appears to have held, but the trial was not designed to prove it. The strongest evidence is the supervisors’ ratings: in Haar’s surveys, team performance did not change between the five-day and four-day periods. That is a meaningful result for a firm that gave up a fifth of its working time. It is also a rating by managers who knew what the trial was for, collected over eight weeks, with no comparison group working a normal week.

The staff measures are clearer about what they show. Work-life balance and stress were self-reported in surveys before and after the trial, with extra data from the end of the previous year. Haar compared the post-trial job attitude scores with New Zealand data from more than 6,000 employees and said they were the highest he had seen. That tells you how staff felt about the change, which matters, but it is not a measure of output.

The bigger numbers came from elsewhere. Engagement rises of 20 to 40 percent and a 35 percent fall in non-work internet use were company figures, and the 20 percent productivity rise that circulated in the press had no stated source; none of them was part of the researchers’ published findings. It is fair to report them, as long as you say who measured them. It is also fair to note that the trial ran under intense media attention, with a founder who had publicly committed to the idea and staff who knew a permanent day off depended on the result.

None of this means the trial failed. A short, company-wide test with independent researchers was a reasonable first step, and far more than most employers do before changing a policy. But it answered the question “can this work here, for a while, under close attention?” more convincingly than the question “does a four-day week raise productivity?”

What did later research add?

The most important follow-up came from one of the original evaluators. In 2022 Helen Delaney and C. Casey published a peer-reviewed case study in Employee Relations of a management-led four-day week trial at a medium-sized New Zealand financial services company, based on focus groups and interviews with 45 employees. They found the shorter week was popular and brought real individual benefits. They also found that performance measurement, monitoring and productivity pressure intensified, and that the collective, pro-social concerns of labour-led campaigns for shorter hours were absent.

That matches signals in the original findings. The researchers’ summaries noted that many employees described the reduced hours as a gift and a privilege, and said they were willing to be available for work on their day off. Those are signs of goodwill. They are also exactly the exchange Kelliher and Anderson described: people who value a flexible arrangement often repay it with extra effort, and the extra effort rarely shows up in a satisfaction survey.

Kelliher and Anderson found a second pattern that is relevant to any 100:80:100 design. Reduced-hours workers whose workload did not shrink with their hours described intensification that was imposed rather than chosen: working late, working on non-scheduled days or staying reachable. In a model where the day off depends on hitting full output, the line between a smarter week and a compressed one depends on whether work was really removed.

Goal-setting research adds a final caution. Locke and Latham found that goal effects are weaker on complex tasks, where people need time to learn better methods, and that very hard goals tied to a reward can backfire when people see they will miss them. A trustee company’s routine work may suit a fixed output goal well. Complex or unpredictable work, such as a difficult estate, may not. Microsoft Japan’s 2019 trial, covered in an earlier case, showed a similar gap between changed habits and proven output.

How would a modern team decide?

A modern team would start by defining output before the trial begins, role by role, in measures the business already trusts: files closed, response times, error rates, client complaints. Perpetual Guardian asked staff to help define productivity, which built commitment. Writing the measures down in advance, and publishing them, would make the result harder to dispute either way.

It would build in a comparison. That could be a staggered start, with some offices beginning a quarter later, or a longer trial that covers busy and quiet seasons. Eight weeks in March and April cannot show whether output holds at year-end, during staff absences or when a large estate lands unexpectedly.

It would measure the hidden costs as well as the benefits. Alongside satisfaction and stress scores, it would track hours actually worked, contact on days off and how often the day off is lost or moved. If a growing share of people are taking calls on their free day, the policy is borrowing effort rather than saving it.

Finally, it would cut work, not just time. The changes that made Perpetual Guardian’s week possible were shorter meetings, fewer interruptions and automated manual tasks. A modern team would review workload before the trial and keep reviewing it, so that a four-day week stays a redesign of the work rather than five days squeezed into four.

The manager's playbook

1. Define output before cutting timeAgree role-level measures of output in advance, using numbers the business already trusts, and publish them before the trial starts.
2. Remove work, not only hoursReview meetings, approvals and manual tasks with each team before the trial, so the shorter week has room to absorb the load.
3. Keep a comparison groupStagger the start between offices or teams, so you can separate the policy’s effect from seasonal swings and media attention.
4. Track the hidden costsMeasure hours actually worked, contact on days off and how often the day off is lost, alongside satisfaction scores.
5. Use outside evaluators and say who measured whatIndependent researchers add credibility; label every figure as researcher-measured or company-reported when you share results.

Questions people ask