Skip to content Skip to navigation

Research & Ideas

Search this site

Lost in Translation: Can Silicon Valley Export Its Best Practices?​

A professor of management science and engineering explains why it's not so easy.

A tech startup in Vietnam | REUTERS/Kham

It's no exaggeration to say that certain business practices are so ingrained in Silicon Valley that they are tantamount to unquestionable management truths.

Open office spaces, ideally with the rugged look of warehouse lofts, are now a design staple for any company that considers itself at the cutting edge.

The agile approach of constant reinvention, with product cycles that last only 90 days, is a benchmark of high-growth competitiveness.

And what could be more hallowed than user-centered design, in which engineers and designers solicit ideas from users and incorporate them at every step in the design process?

As it happens, however, these quintessentially modern practices don’t necessarily transfer well between different cultures. It isn’t that software engineers in China or India are less innovative or dynamic than those in Mountain View or Menlo Park. It’s that people from different cultural contexts bring different worldviews to what may seem like the same kinds of work and make sense of these practices in vastly different ways.

For Pamela Hinds, co-director of the Center for Work, Technology and Organization at Stanford School of Engineering, this is an increasingly urgent issue that can trip up the most well-intentioned multinational corporations.

On the one hand, companies function better across borders when their far-flung business units share the same best practices. On the other hand, some ideas get lost in translation. Assumptions that may seem natural to American entrepreneurs, such as the readiness to “fail early and fail often,” can seem unfathomable to counterparts in Shanghai or Mumbai.

To find out how practices do and do not transfer well from one culture to another, Hinds applied the tools of ethnography to study what happened when a major technology company launched similar “innovation centers” in the United States, China and India.

Each of the new centers aimed at re-creating a Silicon Valley ethos. To encourage a feeling of independence, each center was built away from the company’s main local campus. People who had worked in cubicles suddenly found themselves together in open warehouse-style spaces. They were told to incorporate feedback from users at every step in a product’s design. They were put on 90-day product cycles.

The Stanford researchers spent 20 months at the three sites, studying how workers in each country interpreted the practices with their own constellations of logics. The differences were often striking.

Take the comparatively uncontroversial idea of an open-space office with the rough, unfinished look of Silicon Valley startups: concrete floors and whiteboard walls. At the American center, workers seemed to thrive. They engaged in impromptu conversations and followed up with discussions at each other’s desks. They held meetings in social areas, and reported being able to set up those meetings far faster than had been possible on the main campus. Overall, the American workers reported greater opportunities for collaboration and creativity.

In China, by contrast, people continued to work silently at their individual desks. The only sound in the office, workers reported, was often the humming of the air conditioning. Indian workers had an even more unexpected reaction, actually isolating themselves more from each other after they moved out of traditional office cubicles to the open-space layout. When the Indian engineers had worked in the cubicles, the researchers reported, they had been highly social and interactive. In the new open-space center, however, they increasingly retreated into themselves by reserving time in “quiet spaces” and meeting rooms. For the Indian workers, the semi-privacy of cubicles had provided a foundation for vibrant interaction in the common areas.

Similarly, the mandate to embark on 90-day product cycles produced very different results in different nations. At the U.S. innovation center, people saw the rapid cycle as a way to spur flexibility and be more entrepreneurial. Workers wore multiple hats and routinely took on tasks outside their formal job descriptions. They placed a high emphasis on exploring new ideas and on quickly killing off the ones that didn’t seem to have much promise. They talked about the “courage to fail” and saw failure as a healthy and even essential part of the innovative process. Intriguingly, the U.S. workers didn’t treat the 90-day deadlines as set in stone. In fact, they sometimes let some product cycles run significantly longer.

By contrast, the engineers in China saw the 90-day deadlines as absolute, and primarily as a way to motivate harder work. But they didn’t try to become more flexible or experimental. Indeed, they worried even more than before about rushing out with new software that might flop.

“If we fail, the customer would be angry and we would lose him,” one engineer said. The project “just cannot go wrong."

The Chinese and Indian teams also had markedly different attitudes about “user-centered design.” At the U.S. innovation center, the engineers and designers were entirely at ease with systematically collecting feedback from customers and with tweaking their projects in response.

Chinese engineers, by contrast, actually recoiled at the idea of asking users for advice. They described the users as “indecisive” and “inarticulate,” and they even worried that users would be annoyed at being asked for their opinions. The Chinese engineers also disliked the constant stream of design changes, which are a natural result of getting feedback from users. “I know the plan always changes,’’ one engineer told the Stanford team, “but we don’t like that.”

Hinds and her colleagues don’t argue that one mindset is superior to another. Their goal is to understand how people from different cultures recontextualize a practice so that it makes sense to them. If companies fail to understand those issues, she warns, seemingly sensible practices in one nation can backfire in a different one.

At the U.S. center, they argue, workers brought an “entrepreneurial” logic – a belief in creativity, a tolerance for risk and uncertainty, a willingness to fail and a strong belief in listening to users. Chinese software developers, by contrast, called upon an “engineering” logic and were influenced by assumptions that grew out of being embedded in a centrally planned economy. Indian engineers had a market orientation, but with an emphasis on aligning with the values of the community – assumptions tied to “cooperative capitalism,” as Hinds and her colleagues put it.

Hinds cautions that none of this means that multinational companies should give up on a common array of best practices. What it means, she says, is that companies need to understand the different “constellations of logics” that people from different cultures bring to work.

One key insight: Focus less on having people mimic the specifics of a practice than on understanding the underlying goal of that practice. “If you want people to be more collaborative, you might have open office spaces as a means of getting people to interact more with each other,” she says. But as Indian engineers demonstrated, there are times when the best way to foster collaboration is with old-fashioned cubicles.

Get Updates from Stanford Engineering