Analysis: return-to-office mandates and hybrid deals are often argued over with anecdotes. Here is what the best studies on working from home productivity actually show.
Is working from home good for productivity? The honest answer is that it depends on how much of the week is spent at home and what kind of work is being done. The strongest evidence suggests hybrid working, with two or three days at home, leaves performance broadly unchanged while making staff less likely to quit. Fully remote work is more mixed: some careful studies have found clear productivity losses, particularly for routine tasks and for junior staff who learn by working alongside colleagues. Working from home productivity is not one number, and anyone who claims otherwise is simplifying.
Why the working from home productivity question is hard to answer
Productivity is easy to define in theory (output per hour of work) and hard to measure in most office jobs. A call centre can count calls handled. A policy team or a design studio cannot easily count its output. Many surveys therefore ask workers or managers how productive they feel, which is useful but open to bias in both directions.
There is also a selection problem. People who choose to work from home may differ from those who do not, in age, seniority, caring responsibilities or simply ability. Comparing the two groups does not reveal what working from home does to the same person. That is why randomised experiments, in which an employer assigns people to home or office by chance, carry particular weight in this debate.
How many people in Britain work from home?
Hybrid working is now a settled part of the British labour market, but it is concentrated among better-paid and better-qualified workers. According to the Office for National Statistics, 28% of working adults in Great Britain were hybrid workers between January and March 2025. That rose to 41% of workers with a degree, compared with 4% of those with no qualifications, and to 45% of those earning £50,000 or more, compared with 8% of those earning under £20,000.
Employers have been pulling back slightly. The CIPD’s 2025 survey of UK organisations found that 74% had hybrid working in place, down from 84% in a similar survey in 2023, and 65% now require a minimum number of days in the workplace.
The case that working from home helps
The biggest randomised trial of hybrid work
The most cited study comes from Stanford economist Nicholas Bloom and colleagues, published in Nature in 2024. At the Chinese travel company Trip.com, 1,612 graduate employees in engineering, marketing, accounting and finance were randomly assigned either to work in the office five days a week or to work from home two days a week for six months.
The hybrid group reported higher job satisfaction and its quit rate fell by about a third, with the biggest drops among women, non-managers and people with long commutes. Over two years, there was no difference between the groups in performance reviews or promotions. The company extended hybrid working to all staff, citing savings on recruitment and training.
Commuting, retention and what employers report
Supporters also point to the time and stress saved by not commuting. Dr Darja Reuschke of the University of Birmingham analysed data on more than 16,000 UK employees from the Understanding Society survey and found that satisfaction with the commute had the strongest link to job satisfaction of any factor she examined. For women, having some control over where they worked was associated with higher job satisfaction. She acknowledged that job satisfaction is not the same thing as productivity.
Many employers agree that the arrangement works. In the CIPD’s 2025 survey, 41% of employers said more home or hybrid working had increased productivity or efficiency, against 16% who said it had reduced it.
The case that working from home hurts
Fully remote work and routine tasks
The best evidence against home working comes from a randomised experiment with data-entry workers in India by economists David Atkin, Antoinette Schoar and Sumit Shinde, published as a US National Bureau of Economic Research working paper in 2023. Workers assigned to work from home were 18% less productive than those assigned to the office. About two-thirds of the gap appeared on the very first day, and the rest came from faster learning in the office.
Strikingly, the workers who preferred home working lost more productivity when at home (27%) than those who preferred the office (13%). The authors linked part of this to caring responsibilities and distractions at home. That matters for employers who let staff simply choose: the people most keen on remote work may not be the ones who do it best.
The cost to learning and mentoring
A second concern is the long-term effect on junior staff. Research by Natalia Emanuel, Emma Harrington and Amanda Pallais at a large US company found that software engineers who sat in the same building as all their teammates received 23% more online feedback on their code. But there was a trade-off: being in the office reduced senior engineers’ own coding output by 21%, probably because they spent time helping colleagues. “Young workers learn more when they are working near their co-workers,” Professor Pallais told the Harvard Gazette.
This suggests that some of the benefits of the office do not show up in short-term productivity figures at all. They appear later, in the skills and careers of people who were trained well.
Hybrid versus fully remote: where the evidence points
Read together, the studies are less contradictory than the headlines suggest. They broadly point the same way:
- Hybrid working with a couple of home days appears roughly neutral for measured performance and positive for retention, which saves employers money.
- Fully remote working can reduce output, especially in routine, closely measured tasks and where home conditions are poor.
- Junior and newer staff appear to gain most from time near colleagues, while experienced staff may work faster alone.
- Who chooses matters: letting everyone pick freely can produce different results from a planned team rota.
Experts reviewing the Trip.com study stressed the limits of generalising. Dr Paula O’Kane of the University of Otago, commenting via the Science Media Exchange, noted that “every role, organisation and sector is different”.
What the evidence does not show
There is still little rigorous, UK-specific evidence that measures output directly rather than through surveys. The leading randomised trials were run in China, India and the United States, in particular firms with particular jobs. Nor can these studies settle whether home working has helped or hurt Britain’s national productivity figures, which are shaped by many other factors.
The balanced conclusion is that the question “is working from home productive?” is too broad to answer with a yes or no. A better question for any employer is which tasks benefit from being together, which benefit from quiet focus, and which staff need time near colleagues to learn. The evidence supports designing hybrid arrangements around those answers, rather than either a blanket return to the office or open-ended remote work. For a different view on office attendance in the public sector, read our opinion piece arguing civil servants should be asked to come into the office.
Frequently asked questions
Are people who work from home more productive?
Not as a general rule. A large randomised trial of hybrid working found no change in performance, while a randomised trial of fully remote data-entry work found an 18% drop. The answer depends on the job, the person and how many days are spent at home.
How common is hybrid working in the UK?
The ONS found that 28% of working adults in Great Britain were hybrid workers in early 2025, with much higher rates among graduates and higher earners.
Does hybrid working reduce staff turnover?
In the Trip.com trial, quit rates among hybrid workers fell by about a third compared with full-time office staff, which is one of the clearest benefits found in the research.
Why do some employers want staff back in the office?
Common reasons include mentoring and training junior staff, collaboration, and concerns about output in some roles. Research on software engineers found that people near their teammates received more feedback, which supports the training argument.
Sources: Office for National Statistics, Who has access to hybrid work in Great Britain? (11 June 2025); CIPD, Flexible and hybrid working practices (15 July 2025); Science Media Exchange, Expert reaction: working from home flexibility can improve staff retention without impacting performance (June 2024); National Bureau of Economic Research, Working from Home, Worker Sorting and Development (July 2023, revised December 2023); Harvard Gazette, New to office life? Not having an office might hurt (June 2023); University of Birmingham, Is working from home bad for employees’ productivity? (7 February 2025).
