The search strategies used in research have been found to be very important to the resulting nature and degree of innovation. James March (1991) described two strategies firms could use when searching for new ideas or technologies. An explorative search strategy focuses on searching through unfamiliar material in the hopes of a breakthrough that would expand an area of specialty. An exploitive search strategy looks at areas of existing competence to find ways to better use or extract value from current resources. Studies that have built on the importance of search scope in innovations include Nerkar (2003), Gittelman (2003),Rosenkopf and Nerkar (2001), and Katila and Ahuja (2002).
In measuring innovative performance, this work traditionally focuses on patents, the firms’ ‘‘claim’’ on a technology (Cohen et al. 2000; Pakes and Shankerman 1984) and its Innovating knowledge communities incentive to invest in R&D (Trajtenberg 1990). Although they are publicly available, patents are not filed to disseminate knowledge. In contrast, the knowledge communities we study in this paper work primarily by publishing papers on technical subjects in journals or presenting papers in conferences (Garfield 1983, 1988). Citations in papers, like citations in patents, are made when another work had been influential to a discovery, or when a discovery is based on another work (Merton 1965).
Marketing The well-known theory of ‘‘integrated marketing’’ holds that organizations which present a unified and clear message to customers are more likely to perform well (Phelps and Johnson 1996; Schultz et al. 1994). Consistency and dependability in marketing messages is the essence of brand identity, one of the most robust findings in the marketing literature (Haynes et al. 1999).
A strongly focused, sharing collaborative group culture can help gather and triangulate information about a market to maintain a better window into its needs and preferences (Hurley and Hult 1998; Pfeffer 1993; Powell et al. 1996; Slater and Narver 1995). Such a group has several interlocking abilities that allow it to communicate a message to the market in a unified, coherent way (Slater and Narver 1995). The abilities to market products and ideas are not so different, as argued by literature on ‘‘science wars’’ and discussions on how ideas propagate as ‘‘memes’’ (Kogut and Macpherson 2004). Strategies for selling ideas can be in many ways analogous to strategies for selling tangible products (Downs 1957; Hotelling 1929).
Knowledge communities are interesting in this respect because the ‘‘market’’ for a knowledge community is perhaps primarily internal to those within the community and then secondarily—but necessarily—to those beyond the community. This increases the value of a shared internal form of presentation and uniform norms for use of language and methods (McCloskey 1998).
We examine two aspects of knowledge communities: how they use and build on previous knowledge, and how they use language (Abrahamson 1996; Braam 1991a, b). We believe that the way a knowledge community uses previous knowledge will be closely related to the way it has searched its relevant knowledge-space for new ideas (Katila and Ahuja 2002; Nerkar 2003; Rosenkopf and Nerkar 2001). On the other hand, we believe the way a knowledge community uses language will be more closely related to how it communicates internally and externally—how it markets its ideas to attract attention and members (Abrahamson 1996; Powell et al. 1996; Slater and Narver 1995). In explaining differential success, we believe that the existence and stability of community attributes, and their low appropriability, is best explained under a broad resource-based view framework that takes into account interconnected capabilities, norms, and routines. Previous research has shown how important to innovation it is how a knowledge community searches for new ideas (Katila 2002; Katila and Ahuja 2002; Rosenkopf and Nerkar 2001). Literature on search strategies implies that broad search will be more likely to result in innovation (Cohen and Levinthal 1990; Fleming 2001; Fleming et al. 2005; March 1991). Therefore, we believe that a firm which looks broadly in its intellectual landscape will be more likely to identify valuable sources of ideas. S. P. Upham et al.
H1 Innovating knowledge communities that draw from diverse sources of knowledge will perform better.
Broad search, however, will only be useful to the extent that a knowledge community can exploit the good ideas it finds (March 1991). In quickly moving areas of science, therefore, this ability to draw from sources with valuable ideas will signify itself in a knowledge community’s ability to redirect its focus as an intellectual community as its research evolves and new areas of knowledge become more useful (Fleming 2001; Levinthal and March 1981). The broad range of knowledge collected within a tight and interconnected knowledge community, as hypothesized in H1, allows such a community to scan the intellectual and technical landscape quickly and efficiently (Uzzi 2005; Watts 1999). The ability to search broadly within a tight community and beyond it is crucial to finding valuable resources, and the ability to then reposition to take advantage of these resources is also crucial (Fleming and Sorenson 2001; Rosenkopf and Nerkar 2001).
H2 Innovating knowledge communities that are flexible in their use of knowledge will perform better.
It is important for a firm to search and take advantage not only of standard resources but of non-standard ones (Katila 2003 draft; Katila and Ahuja 2002). Particularly where academia and industry mix, the emphasis is on ideas that contribute something innovative (Gittelman andKogut 2003; Henderson et al. 1998; Murray and Stern 2005; Pakes and Shankerman 1984). Therefore, a knowledge community must search in areas that are somewhat unique (Cohen and Levinthal 1990; Drucker 1985; Katila 2003 draft; Schumpeter 1934). Ceteris paribus, the more competitors searching in an area, the less likely it is that undiscovered value can be found there (Porter 1985). Further, it follows that a firm which searches broadly (H1) and is flexible (H2) will more likely be among the first to find valuable untapped resources (Lieberman and Montgomery 1988; Makadok 1998).