Safeguard Global Scrapped Its Four-Day Week: What Went Wrong

Safeguard Global gave its engineers every Friday off in 2023, hoping a four-day week would help the company hire and keep people during the Great Resignation. Three years later, the Austin-based workforce management firm has scrapped the fixed schedule entirely, after discovering that Fridays off did not mean the work actually stopped.

In a recent commentary piece originally published on Fortune.com, chief executive and founder Bjorn Reynolds described what went wrong. The pilot showed why a four-day week needs more than a new calendar. Some employees quietly kept working on Fridays. Others crammed five days of work into four longer days and burned out. And the people who did protect their day off worried they were falling behind, which according to Reynolds was not the culture the company wanted to create.

The pilot that was supposed to reduce burnout

Safeguard Global employs nearly one thousand remote workers across 78 countries and sells employer-of-record and global payroll services. The pilot ran from January to July 2023 and covered a small group: about 20 engineers spread across 26 countries, plus one senior product leader. According to HR Brew, which interviewed Reynolds about the experiment, the goal was recruitment and retention, and early on employees said they appreciated the three-day weekends.

The trouble surfaced gradually as the four-day week trial wore on. Managers began noticing work seeping into Fridays and found themselves reminding people to log off, Reynolds said. The senior leader in the trial typically worked at least part of the supposedly free day, even after Reynolds personally told her to get offline whenever he saw her online. Responsibilities and expectations outlasted the formal schedule change.

Fridays went quiet, but the work did not

Later employee feedback revealed a deeper pattern. Some staff stretched their Monday-through-Thursday days to accommodate the same workload, which led to burnout rather than relieving it. Others simply worked on Friday without telling anyone. A third group followed the policy faithfully and then felt anxious that taking the day off was leaving them behind on deliverables and client communications, according to reporting on the follow-up.

The four-day week had replaced one fixed schedule with another. For a company whose clients need service across time zones and borders, telling everyone exactly which four days to work was not flexibility at all. As Reynolds put it in his commentary, telling people exactly when they have to be flexible is not really flexibility.

From fixed schedules to optionality

Safeguard Global now runs on what Reynolds calls a culture of freedom and choice. Each of its remote employees can choose working hours and days that fit the work. Someone might complete everything in a four-day week, work five shorter days, or split the week around partial days, with the rule that whole teams coordinate on the same rhythm so the day off can actually be enforced.

Performance is measured against three specific, role-dependent outcomes rather than hours logged or digital availability, according to HR Brew. A customer service representative, for example, is evaluated on account growth, customer satisfaction scores, and points of failure. If support tickets spike or resolution times drift past company benchmarks, managers can investigate whether the cause is a client anomaly, a qualitative issue, or an unsustainable personal scheduling pattern.

Reynolds said the shift made him more disciplined as a leader, not less. Giving people freedom requires spelling out exactly what success looks like, which he described as a healthier management conversation than dictating when someone has to be in front of a computer. Accountability, he wrote, has to come from the top: if managers need hours worked to judge performance, the organization probably has not defined its measures clearly enough.

The lesson: schedules are not interventions

Safeguard Global's experience does not prove that four-day weeks fail. A large United Kingdom trial involving 61 organizations found employees reported less burnout and stress while most companies maintained or increased revenue, with nearly all planning to continue, according to the Associated Press. The contrast is the point: the shorter schedule is not the intervention. The intervention is the collection of changes that make a shorter schedule possible.

That distinction applies well beyond four-day weeks. Hybrid policies stumble when companies change where people work without redesigning communication and coordination. New tools disappoint when organizations add them without removing obsolete processes. As one analysis of the case put it, a new policy cannot produce new results if leaders leave the system underneath it unchanged.

Leaders considering a four-day week should first ask what work will stop, which expectations will change, and how coverage will work, the analysis advised. They should also watch for quiet signs that the model is failing: messages sent on protected days, longer working hours, delayed handoffs, and employees who only meet expectations by concealing extra effort. Otherwise the fifth day does not disappear. It just becomes invisible.

The Safeguard Global four-day week story lands in the middle of a wider argument about how work should be measured. Companies such as Amazon, UPS, and Dell have tightened physical attendance rules, while others like Disney have pushed aggressive mandates, as covered in Disney's four-day in-office mandate. At the same time, surveys keep finding that 62% of workers say they are overloaded even as headcount shrinks. Reynolds' answer is that calendars should follow outcomes, not the other way around, and that he explained in his own commentary on the experiment and in interviews where he told HR Brew.