Here is an excerpt from an article written by Anne-Laure Fayard and John Weeks for the Harvard Business Review blog. To read the complete article, check out the wealth of free resources, and sign up for a subscription to HBR email alerts, please click here.

*     *     *

Managers once discouraged, even forbade, casual interactions among employees. To many bosses, chitchat at the watercooler was just a noisy distraction from work. Today we know that chance encounters and conversations on the job promote cooperation and innovation, and companies craft their floor plans and cultures with this in mind. The results have been surprising—and often disappointing.

Consider the experience of Scandinavian Airlines (SAS). In 1987 the company redesigned its headquarters around a central “street” that linked a café, shopping, and medical, sports, and other facilities, including several “multirooms” containing comfortable furniture, coffeemakers, fax and photocopying machines, and office supplies. The new design was explicitly intended to promote informal interactions, and management broadcast the message that employees should find opportunities in the new space for “impromptu meetings” and “creative encounters.”

What happened as a result? Very little. A study of employee interactions revealed that just 9% were occurring in the street and the café, and just 27% in all the other public spaces combined. In spite of the thoughtfulness and good intentions informing the new design, two-thirds of interactions were still confined to private offices. What went wrong?

Common sense, it turns out, is a poor guide when it comes to designing for interaction. Take the growing enthusiasm for replacing private offices with open floor plans in order to encourage community and collaboration. More than a dozen studies have examined the behavioral effects of such redesigns. There’s some evidence that removing physical barriers and bringing people closer to one another does promote casual interactions. But there’s a roughly equal amount of evidence that because open spaces reduce privacy, they don’t foster informal exchanges and may actually inhibit them. Some studies show that employees in open-plan spaces, knowing that they may be overheard or interrupted, have shorter and more-superficial discussions than they otherwise would.

Both sets of findings are correct. Open floor plans, or indeed any type of design, can either encourage or discourage informal interactions, depending on a complex interplay of physical and social cues. Over the past 12 years we have conducted nine studies of the effects of design on interaction, looking at organizations in the United States, Europe, and Asia. We surveyed the extensive literature on the subject and interviewed dozens of managers about their office redesigns. The sum of our research reveals that a space may or may not encourage interaction depending on how it balances three dimensions, or “affordances,” that have both physical and social aspects: proximity, privacy, and permission. (For more on affordances, see the sidebar “The Signals Design Can Send.”)

The most effective spaces bring people together and remove barriers while also providing sufficient privacy that people don’t fear being overheard or interrupted. In addition, they reinforce permission to convene and speak freely. These requirements, we’ve found, apply just as readily to virtual spaces as to physical ones, although their virtual manifestations may be quite different. In either setting, getting the balance wrong can turn a well-meant effort to foster creative collaboration into a frustrating lesson in unintended consequences. Although no formal studies of the reasons for the design failure at SAS were done, it has all the earmarks of such an imbalance—and should serve as a cautionary tale for any company contemplating a redesign.

The Properties of Proximity

People often assume that proximity is purely a function of physical factors: how far employees are from one another or how close they are to a break room. And distance is important. The MIT organizational psychology professor Thomas Allen famously discovered that the frequency of workers’ interactions in an R&D complex he studied declined exponentially with the distance between their offices—an effect popularly known as the Allen curve. Even when they were in the same building, researchers on different floors almost never interacted informally, he found.

To continue reading, subscribe now or purchase a single copy PDF.

Already an online or premium subscriber? Sign in or register now to activate your subscription.

*     *     *

Anne-Laure Fayard is an assistant professor of management at the Polytechnic Institute of New York University. John Weeks is a professor of leadership and organizational behavior at IMD in Lausanne.