The Ladder

Working From Home Isn’t for Everyone

And if you think you’d love it, there’s a good chance you’re wrong.

Until you try working from home, it’s difficult to tell whether you’ll like it and be good at it.

Photo illustration by Slate. Photos by Thinkstock.

Shortly before I graduated from college, I was hired to do a job that required me to work from home. I was glad to have a job at all, given that the Great Recession was in full swing, and I was particularly excited to have this job, which involved developing recipes—something I liked doing. I bought one of Ikea’s cheapest desks and set it up in the corner of my newly rented studio apartment, determined to work regular hours in a space devoted exclusively to my job.

Within months, I was miserable. I had abandoned the uncomfortable desk for my comfortable futon, erasing all boundaries between work and the rest of my life. I procrastinated during the day—it was so hard to ignore the dishes in the sink or my favorite websites—and then made up for lost time at night. Some days I didn’t shower or go outside, until eventually I grew disgusted with myself and forced myself to take a gloomy walk. I was very bad at working from home.

Could I have done anything to make my working-from-home adventure less of a slog? There is tons of advice out there about working from home effectively, and it definitely is effective for some people. Graham Davis, for example, started working from home as the head of PR for custom-printing firm in March, and he’s adjusted his working environment to overcome certain roadblocks. “At first I was trying to work out of my living room and just found myself being so distracted,” Davis told me. So he set up a home office in a spare bedroom and decided to work only during normal business hours. He’s incorporated a morning walk into his daily routine, and then he showers and changes into work clothes instead of staying in his pajamas, and now he enjoys working from home. “I kind of knew that I would like this, but until I actually did it I didn’t know for sure,” he said.

If you’ve fantasized about skipping the commute and working out of your house, you probably think you’ll be efficient and productive, like Davis. But a recent study suggests you have an equal chance of feeling distracted and isolated, like me. It turns out that people are very bad at predicting whether they’ll be good at working from home—and that some people simply aren’t cut out for it, no matter how well they structure their workspaces and time.

The study, published in the Quarterly Journal of Economics earlier this year, is the first randomized controlled trial of working from home in the management literature. Overall, it vindicates the practice of allowing employees to work from home: Employees in a call center for the travel agency Ctrip in Shanghai were asked if they would like to try working from home four days a week. Some volunteers were disqualified because they hadn’t been at the company long enough, didn’t have private rooms in their homes, or didn’t have Internet access. Half the rest were randomly selected to work from home four days a week.

On average, the home workers’ performance improved 13 percent, because they worked longer and took more calls per minute, and their attrition rate fell by half compared with the control group. They also reported increased job satisfaction. Sounds great, right?

But despite the overall gains of the working-from-home group, half of them decided to go back to working in an office after the nine-month experiment ended—mostly because they said they felt lonely when they worked at home. (Workers whose performance suffered when they worked from home were more likely to decide to return to the office.) “We were astounded that half of them decided to come back into the office, despite the fact that they were saving an hour and a half a day on commuting, and on average earning quite a bit more at home because they were performing so much better,” says Nicholas Bloom, the Stanford economist who was the lead author of the study.

It’s worth emphasizing that those Ctrip workers had lots of factors working in their favor. They volunteered for the experiment, which means they were presumably motivated to work from home. Ctrip actually installed workstations in employees’ homes—essentially creating home offices for them. And the passive nature of their work made it hard for them to shirk their duties: They just had to put on their headsets and take calls. “It’s just really hard to tell what it’s like until you try it,” says Bloom. “We’re not very good at predicting what it’s going to be like working at home for four days a week for nine months. It’s just something that’s way beyond our experience scope.”

When Bloom and his colleagues tried to find correlations between working-from-home success and other factors—how far the employees lived from the office, whether they were married, whether they had kids—they didn’t find any significant relationships. In other words, it wasn’t just the Ctrip workers who couldn’t predict whether they’d do well working from home—the economists, with reams of data at their fingertips, couldn’t do it either.

What does this mean for employers who might like to let their employees work from home—and for individuals thinking about taking jobs that allow or require it? Ultimately, after the nine-month trial period, Ctrip allowed all the workers in the call center to choose whether to work mostly from home or to work in the office, and their overall productivity rose by 22 percent—because now the people who’d discovered they hated working from home were back in the office. So giving employees the option of working from home can do wonders for the bottom line. (This conclusion has support from other studies, too: Allowing employees to work from home part-time is positively associated with profitability, productivity, and good management practices, whereas forcing them to come into the office is associated with being Yahoo.)

Bloom recommends that managers who are interested in allowing employees to work from home institute a trial period without telling employees that their performance is being evaluated. “For example, say that because of the bad weather in January and February, we’re going to have two months of winter working from home,” said Bloom. If things go well, allow people to continue working from home up to a few days per week—but keep an eye on their performance to make sure they’re not floundering.

For managers considering whether to hire people who will work exclusively from home—or entrepreneurs hoping to save money on office space—it’s important to hire carefully., where Davis works, has no central office—everyone works from home—and its CEO, Joe Golden, is very cautious when interviewing to make sure that potential hires can actually work well at home, instead of just thinking they can. “I’ll ask questions about: Have people worked from home before? And in what capacity and how recently? Why do you want to work from home? Do you have a good office space you can use?” says Golden. Bloom endorses this kind of careful hiring—“or you can just put them on a trial period—that’s the other way to do it,” he says. If you don’t hire carefully or put people through a trial period, the fallout can be rough—Golden says that he once fired an employee who “just wasn’t really responding and wasn’t getting their work done” (although he notes that he can’t say for sure whether the person would have done better in an office environment).

Bloom’s study suggests that people have very little self-knowledge when it comes to their aptitude for working at home—but is it possible for people who struggle with working from home, like me, to transform into people who flourish, like Davis? Maybe. There’s no hard data on whether people can willfully transform themselves into successful telecommuters, but Bloom says that, personally, having children around makes him less prone to loneliness—the Ctrip telecommuters’ biggest complaint. “My sense is, as you get older, you find it easier to concentrate at home,” he says, “but I don’t have any hard evidence on that. But I think it’s quite reasonable that 19-year-olds get itchy being on their own for too long, whereas older individuals don’t.”

So maybe there’s hope for me, and someday I’ll be happily working away in my apartment. Until then, as I bike through traffic and head into the office every morning, I’ll try to be grateful that I have somewhere to go.

Need career advice? Got a problem at work? Email