A massive new study holds some startling insights for the age of remote work

After 3 ½ years of remote and hybrid work, we now know this assumption was wrong. Some studies, in fact, have found that working from home actually allows white-collar professionals to get more done, not less.

But productivity is only one measure of labor — and a limited one, at that. The quantity of work people complete on a day-to-day basis is a very different thing from the quality of that work. Output per hour tells us nothing, for instance, about innovation — whether teams are coming up with new ideas that have the potential to turn into game-changing products and services down the road. Productivity is important for running a business in the short term. Innovation determines a company's ability to survive, and thrive, in a changing economy over time.

So: Does working from home stifle collaboration and hamper innovation? Many executives believe it does, but there haven't been many good studies to prove it. After all, remote work has been taking place on a mass scale only since 2020, and new ideas take years, if not decades, to develop into real-world applications. It's hard to measure something that hasn't happened yet.

Now, though, a massive new study published in the journal Nature has shed new light on the effect of remote work on innovation. At the center of the study is a clever workaround. Even though remote work is a relatively new development in corporate settings, scientists and inventors have been collaborating over long distances for decades. So researchers at Oxford University and the University of Pittsburgh pored 20 million scientific studies and 4 million patent applications from around the world over the past half century. Who had the better track record of arriving at breakthroughs — teams that worked remotely, or those that collaborated in person?

What the researchers found is striking, and has big implications for employers — especially in Silicon Valley, where businesses can become irrelevant in the blink of an eye. Collaborating in person, the study found, really did produce more breakthroughs than remote work. And the farther away team members were from one another — even if they were in the same time zone — the worse chance they had of producing work that was groundbreaking. Teams located in the same city, for example, were 22% more likely to produce innovative patents than teams spread out by several hundred miles or more — and 27% more likely to produce pioneering insights in scientific papers. IRL beat WFH by a mile.

"I wouldn't say that all companies should go back to being fully on-site," says Carl Benedikt Frey, an economist at Oxford who was a coauthor of the paper. "But if you think of it purely from the perspective of trying to develop breakthrough technologies, you should probably be on-site as much as possible."

The researchers then combined those D scores with the locations of the collaborators on each paper and patent. That allowed them to calculate the probability that a set of collaborators would produce groundbreaking work, based on how far away everyone was from one another. The closer the team, the more innovative they were likely to be.

The big question is why. If you ask the executives out there who are demanding that their employees return to the office full time, they'll probably opine on the mysteriously magical quality of conversations at the watercooler that somehow makes everyone more creative.

And it looks like they're right — sort of. Frey and his coauthors looked at their data on scientific papers to see what they could learn about the specific contributions made by each collaborator. Among teams that worked in person, many of the collaborators participated in the initial stages of conceiving the research. On remote teams, by contrast, the more established collaborators tended to come up with the original idea on their own. Then they delegated the technical tasks of the study to their less-established peers. Apparently, having better, more collaborative interactions about the early, big-picture thinking — everybody gathered around the watercooler, as it were — actually does lead teams to pursue more novel ideas.

This difference between remote work and in-person collaboration probably makes sense to anyone who's tried to articulate a half-baked thought on Slack. When you set out to write an article, Frey says, "the very first conversation you have about it is usually not that crisp. It's slightly embarrassing. If you sit in the same room as somebody and something occurs to you, you'll probably just turn around and ask a question and get the input and feedback. But if it requires you to pick up the phone or send an email, then you might not even bother."

Now, before companies start racing to reorganize their workplace policies based on this single study, a few words of caution. For starters, the work of scientists and inventors is pretty different from that of a typical office employee. And just because remote collaboration didn't work for innovation in the past doesn't necessarily mean it won't work in the future. Our tools for remote collaboration have improved dramatically over the past few years, and they'll get even better in the years ahead, now that there's a much bigger market for tools that facilitate remote and hybrid work.

But the study's findings — given the remarkable sweep of the data it examined — do suggest some guidance for companies in the age of remote work. First, before you start hauling your employees back to the office, you need to figure out how much your business relies on innovation versus incremental improvements. Do you succeed by revolution, or by evolution? The vast majority of work in professional jobs is steady and unflashy — and there's nothing wrong with that. "If you're an established player that already produces a product," Frey says, "then you have greater benefit to investing in technologies that support scale and incremental improvements."

Moreover, even at the most pioneering tech companies, not every worker is involved in the innovative core of the business — the part of the operation that's trying to come up with exciting new things. Yes, the product teams will probably stumble upon better ideas if they're together in person. But the people in roles like accounting, HR, and IT are there to support the innovators, not to innovate themselves. So they should be fine working from home. That's one reason companywide RTO mandates make little sense. Organizations would be smarter to implement policies based on specific teams and functions, not treat everyone the same.

Even for companies that want to let their entire staff work from home, the study's findings offer some hints on how to avoid losing out on innovation. Rather than letting the most senior bosses come up with ideas for their underlings to execute, look for ways to get everyone to participate in the brainstorming — to create a remote version of the office watercooler. People can be shy about sharing their wacky thoughts on Zoom. But maybe some of those ideas will prove to be a little bit genius. The goal is to get the best of both worlds — the cost savings and recruiting advantages of remote work, and the innovative edge of an in-person office.

Striking that balance is vital not just for individual businesses, but for all of us. Sure, slow and steady improvements are what keep our economy humming along on a day-to-day basis. But to meaningfully improve our standard of living over time, we need groundbreaking discoveries every once in a while. That's why economists are so obsessed with technology, and it's why so many of them are worried about America's recent slowdown in productivity. The pessimists posit that humanity has simply run out of good ideas. But Frey's study suggests that the rise of remote work — facilitated, ironically, by all the breakthroughs in communication technologies over the past few decades — might be partly to blame for our slump in technological innovation. The farther apart we work, the farther away we get from coming up with the next big thing.

"We can improve a horse carriage as much as we want," Frey says, "but unless we build an automobile, we won't revolutionize personalized transportation. We really need these radical breakthroughs — and we're not getting enough of them."

Aki Ito is a senior correspondent at Business Insider.