Innovation is often seen as the lifeline of the modern firm, which has to survive and move ahead in a world of global competition and rapid technological change. “Continuous innovation is required just to keep up with your competitors, never mind lead the pack,” says Lori Rosenkopf, a professor of management at Wharton.

One way organizations innovate is by absorbing new ideas, insights and expertise from outside in an attempt to combine existing and new knowledge. But recent research shows that firms find it difficult to reach for new knowledge beyond their immediate technological and geographical contexts for a variety of reasons – economic, technological and cultural. As a result, companies tend to limit most searches for new knowledge to other companies operating in similar areas of technology or which are located nearby in terms of geography.

This situation – of obtaining new knowledge by conducting “local searches” – has its advantages. For one thing, it means that the cost of searching for new knowledge remains relatively low. In addition, it helps ensure that organizations are more attuned to absorbing familiar types of knowledge. But it is not enough for companies to search for new knowledge only from peers in the same technological and geographic spaces. Local searches tend to block out novel and more distant knowledge. Both are essential for an innovating firm. In other words, companies that want to be truly innovative must be willing both to exploit local sources of knowledge and explore distant ones.

But how can companies reach beyond existing contexts in search of new knowledge? Rosenkopf and Paul Almeida of Georgetown University’s McDonough School of Business raise that question in a research paper titled, “Overcoming Local Search Through Alliances and Mobility.” Says Rosenkopf: “Searching locally is a natural tendency for all firms – similar context leads to similar patterns of communication, similar culture and opportunities to interact. All these things make it easier for technical people to develop stronger relationships and trust one another.”

For example, consider the semiconductor industry. Using patent citation patterns in this business, the authors show both the geographic and technological localization of knowledge. Other studies have demonstrated how every Japanese semiconductor firm (save one) had patents in similar areas. The same trend is visible in other parts of the world. Cases in point: Information flows within regional clusters in Italy and Silicon Valley.

The authors suggest two ways that companies can reach beyond local technological and geographic barriers – through the formation of alliances and the mobility of inventors. Both strategies can serve as bridges to new contexts and help spur innovation.

Rosenkopf and Almeida point out that by forming alliances and tapping into the mobility of inventors, companies can absorb new knowledge. They offer interesting examples where companies or industries have broken out of their local shackles.

The Japanese firm Matsushita accomplished a technological transition though the strategic use of alliances. “The underlying idea is that companies should consider entering into alliances with other companies that are unlike them,” says Rosenkopf, citing Microsoft and Cisco Systems as examples of firms that have done this well. “Such alliances help companies create relationships and get into areas that they might otherwise not have considered.”

Mobility of knowledge workers also provides a strong impetus to fresh ideas and innovative practices. During the early days of the Korean semiconductor industry, for instance, the practice of bringing U.S.-educated and U.S.-employed Koreans back home did a lot to strengthen the semiconductor industry in Korea, the authors note.

Rosenkopft and Almeida argue that “the occurrence of alliances or mobility between firms in a similar context merely duplicate pre-existing relationships but offer little added value to the firm. Distant technological and geographic contexts may offer access to new and unique knowledge.” Companies need to use ways to acquire knowledge so as to fill the holes created by their immediate technological and geographical contexts.

Rosenkopf points out that these research findings have important implications for business executives. “It shows that they should cast a wide net when thinking about possible alliance partners,” she says. “In considering an alliance, they should look at its larger, knowledge-based benefits. It is easy to keep returning to the same partners and allies because of your past experience with them. It is important, however, to invigorate the mix. That can happen when you make uniqueness and non-redundancy important considerations in choosing an alliance partner.”

Rosenkopf and Almeida are now following up this research with two new projects. “In a companion piece with Paul Almeida and Gina Dokko (a Wharton doctoral student), we found that the larger the firm, the less effective it is at knowledge-building through informal mechanisms like mobility and regional networks. We are also planning to do more detailed analyses of how these semiconductor firms are positioned in various knowledge networks of alliances and mobility,” says Rosenkopf.