It’s an old debate: Some cite research suggesting that remote workers earn more, quit less, and are more productive than their office-dwelling counterparts. Others point to evidence that workers at home are less productive and less innovative than workers who labor shoulder-to-shoulder.
Which camp is right? Probably both. And neither. There’s actually only one right answer to the question of whether employees work better at home or in the office, says Ben Waber, the CEO of the workplace analytics company Humanyze and a visiting scientist at MIT: It depends.
“Like a lot of debates we have in society, we tend not to be satisfied when the answer is, ‘it depends,’” Waber says. “But that is absolutely the answer.”
While productivity gurus often take an absolutist approach, arguing for remote work or against it, much of the research that they cite doesn’t attempt to examine remote work in general, but rather its impact on particular types of work within organizations that have specific characteristics.
A European retail bank that hired Humanyze to analyze its office layout, for instance, found that sales teams that spent time interacting in person outperformed those who worked remotely. That appears to contradict an often-cited 2014 study by Stanford researchers that looked at how working from home impacted employees at a Chinese travel agent’s call center. The study found that employees at home were on average 13% more productive, making more phone calls and spending more time on the phone.
But the circumstances of the two workspaces were very different. Members of the sales team, Waber hypothesizes, benefit from learning how others do the job better. In-person, an improvement one person makes is more likely to be shared with others.
The call center employees, meanwhile, had a job that typically doesn’t involve iterating on new ideas or working in teams.
“It’s not a general conclusion that working from home is a good idea,” says John Roberts, one of the paper’s authors. “There are many more opportunities to improve both productivity and employee satisfaction by allowing people to work from home. It’s not a panacea. It’s not something you should do with all jobs.”
Research has explored many other variables that may influence the impact of remote work, including:
- The portion of their work week that workers telecommute: Working from home for one day a week may have a different impact than working from home full-time. Some studies have found that job satisfaction is highest among workers who telecommute a moderate amount.
- The culture of the organization in general: Job satisfaction among telecommuting workers may be influenced by the same factors that influence job satisfaction in the office, things like feedback and relationships with co-workers and supervisors. Flexibility and autonomy have also been associated with job satisfaction among remote workers.
- Teamwork: In one study, workers with high levels of dependence on others to do their work tended to say they were less productive as remote workers.
- Innovation: Face-to-face interactions have been associated with creative work. A Harvard study found that researchers who worked in close physical proximity produced more impactful papers.
“In general, if you were doing a job that is very external-facing—you’re a sales person, you’re a journalist, you’re coming up with a new idea for a research project—Working remotely tends to be pretty good,” Waber says. “If you’re working collaboratively on a project, trying to iterate quickly, in those cases, being in the office more tends to be better.”
Within that conclusion, there are even more nuanced ways to slice this. “Even if I say, ‘in general, coming up with new ideas is more effective at home,’ well, what type of ideas do you want to come up with? If you ask the question slightly differently, what makes people effective actually varies quite a bit.”
The answer can vary not only from employee to employee, but even for the same employee as he or she works on different types of tasks. “If you are part of a team or managing others, you need to be in work most days of the week,” notes Nicholas Bloom, a Stanford professor who co-authored the call center paper with Roberts. “But even so, it’s very helpful to have one or two days off a week.”