The search-transfer problem: the role of weak ties in sharing knowledge across organization subunits

Administrative Science Quarterly, March, 1999 by Morten T. Hansen

In contrast, in weak interunit ties, the necessary interactions for transferring complex knowledge are absent. The interaction between the source unit and the recipient project team is likely to be infrequent. Recipient team members have to interpret and modify the noncodified and dependent knowledge, often in the absence of further explanations, because the source unit is less likely to engage in two-way interactions. When problems occur and questions arise, the source is not immediately available, if available at all. Even if people in the source unit are available, the parties to the transfer have not established a relationship-specific heuristic to communicate knowledge between them, making the transfer effort more difficult. These obstacles take time. They result in statements like "it would have been faster to do it ourselves." The transfer may have become a burden, hampering the progress of the project.

Search and Transfer Combined

The search and transfer arguments developed above lead to opposing effects of weak ties on the benefits of knowledge sharing across subunits. To assess how weak interunit ties affect project completion time through the interunit knowledge sharing process, I use the two knowledge complexity dimensions as boundary conditions to define the search and transfer problem, as shown in figure 1. I contrast weak and strong interunit ties under conditions of low and high levels of knowledge complexity. When project teams end up transferring codified and independent knowledge, weakly tied project teams have an advantageous search position compared with strongly tied teams (lower-right versus lower-left quadrant in [ILLUSTRATION FOR FIGURE 1 OMITTED]). Furthermore, barring a protective source unit, project teams in this situation (whether weakly or strongly tied) do not incur any significant transfer problems, because the knowledge to be moved across subunit boundaries is not complex. The implication is that the search benefits of weak ties should translate into shorter project completion time because of a more efficient knowledge-sharing process. Project teams in weakly tied subunits are likely to find more useful knowledge per unit of search time, or else they are likely to spend less time searching per unit of useful knowledge obtained through the interunit network:

Hypothesis 1: The weaker the interunit ties, the shorter the completion time when the knowledge to be transferred is highly codified and independent.

A different situation occurs when a weakly tied project team attempts to transfer highly noncodified and dependent knowledge from another subunit (upper-right versus upper-left quadrants in [ILLUSTRATION FOR FIGURE 1 OMITTED]). Weakly tied project teams still have an advantageous search position compared with strongly tied ones, but, in transferring highly complex knowledge, they are likely to incur severe transfer problems because of a poor interaction with the source unit.(2) Thus, the net effect of weak ties is likely to be an increase in completion time when the level of complex knowledge is very high: