Companies and government leaders have long tried to recreate the kind of Silicon Valley innovation magic that made Facebook and Google such goldmines. Cities from Cincinnati to Saskatoon have launched startup labs. Companies including Nationwide Insurance, Walt Disney, and Procter & Gamble hold hackathons and set up incubators to nurture new ideas. Traditional companies are even wading into co-working to stimulate interaction.
But, does it take a village to hatch a groundbreaking venture? It can help, but only if you don’t already know everyone there, says new research by Harvard Business School Professor Rembrand M. Koning.
Like party guests who refuse to mingle, entrepreneurs tend to gravitate toward people they already know at these programs, especially if they participate as a team, diminishing the learning potential.
“These institutions may be ineffective channels for peer learning and advice when members are already embedded in existing networks,” Koning and coauthor Sharique Hasan of Duke University argue in Prior ties and the limits of peer effects on startup team performance, published April 11 in Strategic Management Journal.
The risks of sticking to people you know
Past research suggests that nearby peers are more likely to influence performance than those far away, which explains why scientists work side-by-side at lab benches. But can entrepreneurs expect the same benefit at the networking programs to which cities and institutions are pinning their economic hopes?
“If you're trying to make new connections and learn from the other people, I wouldn't bring your posse.”
To find out, Koning and Hasan invited 112 aspiring entrepreneurs to a three-week software development boot camp in New Delhi, India. Before the event, attendees identified the participants they already knew from a roster of names and photos.
Researchers randomly assigned participants to teams and work stations in a large open space and challenged them to create a mobile app for the Indian wedding industry. They tracked participants’ interactions on the boot camp’s Facebook page and through emails sent on the event’s messaging platform. They also surveyed attendees about the advice others provided them.
“We noticed that teams that were near each other didn’t always influence each other,” says Koning, assistant professor of business administration in the school’s Strategy Unit. “That goes against what previous research has said.”
After the first week, participants rated other teams’ prototypes anonymously, without knowing who created them. The three highest-scoring teams were awarded prizes worth 45,000 Indian rupees, or about $790.
While teams that interacted with their high-performing neighbors generated better business ideas themselves, those interactions were more likely to happen among participants who knew few other attendees. The odds of such interactions were 25 percent for participants who didn’t know anyone at the boot camp versus 16.6 percent for those who did. Participants with more prior ties were more likely to miss out on peer effects, or the benefits of learning from those with stronger skills.
“The whole premise of these things is that you’re going to learn from other people and make new connections,” Koning says.
The limits of social engineering
Koning’s research might be sobering to the cities and institutions that invest in startup-focused events and programs in the hopes of stirring innovation and prosperity. Even if a fledgling community emerges from the interactions at these events, the potential peer effects will likely decline over time.
“Let's say you do a hackathon in Delhi with 100 people who don’t really know one another,” he says. “That’s awesome, but now that you have formed a bunch of relationships, it will be harder to engineer relationships the second time around.”
For established companies, Koning’s research raises another potential peril of referral-based hiring. Beyond ending up with a homogenous workforce, managers who hire candidates who already know other employees might have to work harder to get these candidates to collaborate with people they don’t know.
Leave your posse at home
There are no easy solutions, says Koning, who studies the ripple effects that occur when entrepreneurs fail to harness new advice, skills, or marketing opportunities. It’s difficult to shake the gravitational pull of sameness and comfort that people feel in new situations.
In a perfect world, these networking programs would choose only participants who don’t know each other, but that would be nearly impossible in some industries. Plus, deep relationships provide their own benefits, just not if you want to learn new skills.
“Leaders must strike a balance between forming close bonds and disrupting the bonds,” Koning says. For a global company, that might mean convening employees from different teams, locations, and backgrounds for product and strategy discussions.
Koning’s research doesn’t completely dismiss the merits of hackathons, incubators, and other networking efforts. But these programs shouldn’t be the only tool for inspiring peer learning and creativity. If you’re going to participate, you’re better off going it alone.
“If you're trying to make new connections and learn from the other people, I wouldn't bring your posse,” Koning says. “Otherwise, you won’t meet all these interesting people because you’ll be talking amongst yourselves.”
About the Author
Danielle Kost is senior editor at HBS Working Knowledge.
[Image: ColobusYeti]
Related Reading:
Professional Networking Makes People Feel Dirty
How Entrepreneurs Can Turn Lead Into Gold
Will Startup Fishbowl Become the Social Media App for Your Industry?
What do you think of this research?
Share your insights below.