Employee productivity vs hours worked.

What research says.

A widely cited study by John Pencavel (IZA, 2014) looked at detailed records of munition workers during the First World War. It found that output rose in proportion to hours up to a threshold, and above it rose at a decreasing rate. In the data, the bend came at around 49 hours a week, and at the longest hours extra time added very little. The data are historical and specific, but the pattern matches what many managers see today: the last hours of a long week are the least productive.

The health cost of long hours.

In 2021, the WHO and ILO published the first global estimate of deaths linked to long working hours. They found that working 55 or more hours a week led to 745,000 deaths from stroke and ischemic heart disease in 2016, a 29% increase since 2000. Compared with a 35 to 40 hour week, working 55 hours or more was associated with an estimated 35% higher risk of stroke and 17% higher risk of dying from ischemic heart disease. About 9% of the world's population works long hours.

Why extra hours add less.

1. 1
   Fatigue. Energy and attention fall as the day and week get longer.
2. 2
   Errors. Tired people make more mistakes, which take time to fix.
3. 3
   Slower work. The same task takes longer late in a long day.
4. 4
   Less recovery. Short nights and no breaks lower the next day's output too.
5. 5
   Presenteeism. People are present but not working at their best; see presenteeism.

When overtime makes sense.

Short, planned bursts of overtime can be the right call: a launch, a seasonal peak, an urgent fix. The problems start when overtime becomes the normal week. Set a clear end date for any push, give time back afterwards and watch energy and error rates during it. If a team needs overtime every month, the issue is planning or staffing, not effort.

Capacity matters more than hours.

What people produce depends on the energy, focus, recovery and resilience they bring to each hour. thrive research found that people with the least capacity delivered 56% of priority work on time, against 86% for those with the most. Two people working the same hours can produce very different results depending on how depleted they are. Read more about employee capacity.

What to measure instead of hours.

A worked example.

Two teams of ten deliver the same project. Team A works 50-hour weeks for three months, then loses two people to resignation and spends the next quarter hiring and onboarding. Team B works 40-hour weeks, finishes two weeks later and keeps everyone. Over six months, Team B delivers more, at lower cost, with its knowledge intact. The example is illustrative, but the pattern is familiar to most managers: the hours that look productive this month are often paid back later.

Shorter weeks and productivity.

Interest in four-day weeks and reduced hours comes from the same evidence: if the last hours of a long week add little, removing them may cost less than expected, and better rest may raise output in the remaining hours. Results depend heavily on how work is redesigned. Cutting hours without cutting low-value work simply compresses the same load into less time.

Meetings, messages and real working time.

Much of the working week is spent in meetings, chat and email rather than on the work itself. Long hours often reflect this: people do their focused work after the meetings end. Cutting recurring meetings, agreeing response times for messages and protecting blocks of focus time can raise output without adding hours. It also reduces the pressure to work evenings to "catch up", which protects the recovery people need to perform the next day.

What other research suggests.

Studies of call center agents and other workers with measurable output find similar patterns: as daily or weekly hours rise, the time needed per task increases and quality falls. Research on sleep shows that short sleep lowers attention and judgment the next day. Together these findings point the same way as the historical data: beyond normal hours, extra time buys less output, and recovery is part of productivity, not a break from it.

Changing a long-hours culture.

Long-hours cultures change when leaders change what they reward. Praise outcomes rather than late messages, leave on time visibly, schedule messages for working hours and protect recovery after busy periods. Measure energy and focus regularly so the effect of these changes is visible.

Bringing it together.

Hours are an input, not an output. Beyond normal working time, extra hours add less, raise health risks and erode the energy that tomorrow's work depends on. Measure outcomes and capacity, protect recovery and remove low-value work.

What managers can do.

1. 01
   Stop rewarding hours. Praise results, not late messages.
2. 02
   Watch overtime trends. Several weeks of long hours is a warning.
3. 03
   Remove low-value work. Cut meetings, reports and approvals.
4. 04
   Protect focus time. Blocks without interruptions.
5. 05
   Make rest normal. Breaks and leave taken, no replies expected after hours.
6. 06
   Check capacity. Ask about energy and recovery regularly.

Measuring at team level.

Track outcomes, overtime and capacity by team every month. A team whose overtime rises while capacity falls needs relief before output drops.

Recovery is part of productivity.

Sleep, breaks and time off restore the attention and energy that output depends on. Protecting them is a productivity decision, not a perk.

Designing shorter weeks that work.

Shorter weeks succeed when teams first remove low-value work: fewer meetings, fewer reports, clearer priorities. Without that, the same load is squeezed into fewer hours and strain rises.

First step.

Look at overtime, outcomes and capacity for each team this month, and pick one team to relieve first.

Why it matters.

Teams that protect recovery deliver more over a year, lose fewer people and make fewer costly mistakes than teams that rely on long hours.

Hours, remote work and visibility.

Remote and hybrid work make hours harder to see and easier to stretch. Some people work longer to prove they are working. Measuring outcomes and capacity instead of online time removes that pressure and gives a truer picture of performance. See how to improve employee productivity and signs of overworking.

Signs a team is working too long.

1. 1
   Overtime above plan for several weeks. A sustained pattern, not one deadline.
2. 2
   Rising error and rework rates. Fatigue showing in quality.
3. 3
   Messages late at night and at weekends. Recovery is shrinking.
4. 4
   Falling energy and focus scores. The leading signal in pulse surveys.
5. 5
   More short absences. Often the next step after fatigue.