Social networks have been shown to help entrepreneurs acquire the people, capital and knowledge they need for new ventures. But the question remains, how do they form these networks? And what kind of networks offer the best opportunities?

Wouter Stam of HKUST considers this issue in the context of the Dutch open source software industry. During his period of study, 1999-2004, the industry was fragmented and just starting to form links between disparate events and operators. It was therefore possible to see how entrepreneurs built up their networks.

"Industry events such as conferences and business meetings are unique temporary occasions where entrepreneurs can share knowledge, develop industry standards and organize collective action. Bridging between different events was especially fruitful during the early stages of the Dutch industry's development because individual events did not begin to transform into collective action until 2005," he said.

Stam showed how bridging benefited entrepreneurs by looking at the links between attendance at 45 industry events, network building and the subsequent performance of entrepreneurs' firms.

He measured event attendance through record searches (the typical respondent joined two events), and compared this with results on social networking and venture performance that were gathered in two surveys. Social networking was based on how many other respondents nominated a particular entrepreneur in the study as a friend or acquaintance (the average was five times, with a high of 28), and performance was based on sales between 2004 and 2005 and entrepreneurs' own assessments of their firm's progress.

Stam expected that attending "heterogeneous" events in which participants had few shared relationships would improve an entrepreneur's network building and offer more opportunities to act as network brokers, and the results bore this out. Similarly, attending multiple homogeneous events that had few common participants was also helpful to networking building. However, there were some conditions attached.

Those who benefited most from event participation had broader career experience, and they had to be choosy about which events they attended.

"Simply participating in more events didn't increase entrepreneurs' access to structural holes," particularly if they were mainly homogeneous events where the entrepreneur knew many of the participants," Stam said.

"[Such] industry events appear to constrain individual dispositions by limiting the pool of contacts whom entrepreneurs can meet and form ties with."

In terms of the event participation-firm performance link, he found that ventures led by network brokers tended to perform better, thus supporting earlier research on the benefits of forming networks.

"This lends support to the value of organizing networking events where industry participants can meet strangers and bridge structural holes. In particular, entrepreneurs with small and cohesive networks will benefit from attending such events as the findings show it is brokerage that accounts for the performance effects of event participation," Stam said.

"Having said that, event participation is not costless, as homogeneous and redundant events appear to constrain an entrepreneur's network building. So rather than focusing on how many events to attend, entrepreneurs should carefully consider the compositional heterogeneity and bridging potential of the events they consider attending."

That meant targeting events where they did not know anyone, as this would more likely offer an opportunity to be a "critical linking pin" in the network structure of the industry.