The newest U.S. government time-use figures show that about one in three employed people did at least some of their work at home on the days they worked in 2025. This article walks through what the survey measured, who works at home most, and how teams can use the pattern to plan their weeks.
What the 2025 American Time Use Survey measured
The American Time Use Survey (ATUS) is run by the U.S. Bureau of Labor Statistics. Participants describe, hour by hour, what they did on a single assigned day, which is called a time diary. Because people report actual activities rather than opinions about their habits, the survey is a useful check on how work really happens.
According to the Bureau of Labor Statistics, the 2025 results were released on June 25, 2026. About 6,100 people were interviewed. The headline work-location figures apply to employed people on the days they worked:
| Measure (2025, days worked) | Share of employed people |
|---|---|
| Did some or all work at home | 35% |
| Did some or all work at their workplace | 70% |
| Women who worked at home | 38% |
| Men who worked at home | 31% |
These are survey estimates from a sample, so small differences should be read as approximate rather than exact.
Why do the percentages add up to more than 100?
The two headline figures, 35% and 70%, sum to 105%. That is possible because the categories overlap: someone who spends the morning at home and the afternoon in an office counts in both.
A worked example shows what that implies. If 35% worked at home and 70% worked at a workplace, then at least 5 percentage points of employed people (35 + 70 - 100) must have done both on the same day. The true share could be higher, but it cannot be lower. In other words, the question is rarely “home or office?” and often “which parts of the day go where?”
This matters for planning. Teams that talk about remote work as a binary choice miss the large group whose days are mixed, and whose tools, schedules and meeting habits need to work in both settings.
Who works at home, and who does not
The gaps by group are wide. According to the Bureau of Labor Statistics, among employed people aged 25 and over:
- 51% of those with a bachelor’s degree or higher did some work at home on days they worked.
- 19% of those with a high school diploma and no college did so.
That is a difference of 32 percentage points (51 - 19). The gap by sex is smaller: 38% of women against 31% of men, a difference of 7 points.
The pattern reflects what kind of work is possible away from a workplace. Jobs built around a computer, documents and calls can move between locations. Jobs that depend on equipment, customers in person or physical goods mostly cannot. For managers, the lesson is fairness: a flexibility policy that suits office-based staff may leave other colleagues feeling that the benefit is out of reach, so it helps to look for other forms of flexibility, such as predictable schedules or control over shift swaps.
How long are working days, really?
For full-time employed people, the survey found an average of 8.1 hours worked on days they worked. On an average weekday the figure was 8.5 hours, and on an average weekend day it was 5.5 hours.
Illustrative arithmetic: a full-time employee who works five average weekdays logs about 42.5 hours (5 × 8.5). Add one average weekend day and the total reaches 48 hours (42.5 + 5.5). That is a blended average across many people and not a rule for any one job, but it is a reminder that “a standard week” often stretches into weekends, whether or not anyone is at home.
Time-diary data cannot say whether those hours were productive. It records time spent, not output, and the survey does not link location to results. Any claim that home working raises or lowers productivity needs separate evidence.
What the survey cannot tell you
Several limits are worth keeping in mind before using the figures in a policy discussion:
- One day per person. Each participant reports a single day, so the survey describes the population on a typical day and not any individual’s weekly routine.
- No cause and effect. The numbers show who works where, not why, and not what happens to quality, speed or satisfaction.
- A disrupted year. The Bureau of Labor Statistics states that all ATUS operations were suspended from October 1 through November 12, 2025, because of the federal government shutdown, with collection resuming on November 13. Fourth-quarter results were adjusted to represent the full period, and the agency says the effect of the interruption on the estimates cannot be quantified.
- U.S. only. Patterns differ across countries, industries and company sizes, so readers elsewhere should treat the figures as one data point, not a global benchmark.
How should a team plan a mixed week?
Because many days are split between places, the useful design question is which work needs which setting. A simple sorting exercise helps:
- Work that benefits from being together: kickoffs, difficult feedback conversations, onboarding, brainstorming that depends on quick back-and-forth.
- Work that benefits from quiet: writing, analysis, design and careful review. This is the deep, uninterrupted effort described in deep work vs busy work.
- Work that is neutral: routine updates and approvals that can happen anywhere if the process is written down.
Once work is sorted, the calendar can follow. Some teams cluster shared meetings on one or two anchor days and leave the rest of the week free for focused tasks. Others keep a daily overlap window of a few hours and treat everything outside it as flexible. Neither is universally right; the choice should match the work the team actually does.
Making hybrid days run smoothly
Mixed days fail in predictable ways: decisions made in a hallway never reach the people at home, meetings in which half the group sits in a room and half joins by video, and tasks that live in someone’s head. A few habits reduce those problems:
- Write decisions down. A short note after any discussion means location does not decide who knows what.
- Make meetings location-neutral. If one person joins remotely, everyone can join from their own device, or the room can be set up so remote participants are clearly heard and seen. The guidance in how to run meetings that respect time applies directly.
- Keep one shared task list. A single visible system, as outlined in a simple system for managing tasks, means work does not depend on who happens to be in the room.
- Invest in trust. Teams that spread across places do better when expectations are explicit and results, not presence, are what gets noticed, a theme explored in building trust in distributed teams.
Is working from home right for every role?
The survey makes plain that it is not. Seven in ten employed people were at a workplace for at least part of the day in the 2025 data, and fewer than one in five of those with a high school diploma and no college did any work at home. Remote or hybrid arrangements are one tool among several for improving how people work, not a universal answer.
For an individual, a useful test is to track one ordinary week: note which tasks needed concentration, which needed other people, and where each was done most easily. For a team, the same exercise run together often reveals mismatches, such as collaborative work scheduled on days when half the group is away.
Conclusion: plan around the work, not the location
The 2025 time-use figures describe a working world in which home is part of the picture for roughly a third of employed people on a given workday, many days are split across places, and access to home working is uneven. The practical response is to design weeks around the kind of work being done, write down what is decided, and avoid treating any one location as the measure of effort.
This article provides general information only and is not business, employment, legal or financial advice.
Sources: U.S. Bureau of Labor Statistics
