Facts and Figuring: An Experimental Investigation of Network Structure and Performance in Information and Solution Spaces
Executive Summary — How can managers create organizations that bring people together to successfully solve problems? One increasingly popular managerial tactic to improve problem-solving performance is to increase the connectedness, or what academics call clustering, of the organization. Using everything from transparent, open offices to open social collaboration platforms, connecting everyone and everything, the theory goes, will produce better solutions. True or false? In the lab, the authors randomly assigned individuals to 70 sixteen-person organizations—some more clustered than others—and asked each organization to solve a complex problem: divine the who, what, where, and when of an impending terrorist attack (akin to the famous Clue® whodunit game). They did so using a platform not unlike real intelligence problem-solving environments: Through their computers, individuals could search for information, share information with each other, and share theories about the solutions, while the platform tracked all behavior. The results? Connectedness had different effects on the "facts" and "figuring" stages of problem solving. Search for information (facts) was, indeed, more efficient the more connected the organization. But performance in interpreting the information (figuring) to develop solutions was undermined by too much connectedness. The same connections that helped individuals coordinate their search for information also encouraged individuals to reach consensus on less-than-perfect solutions, making connectedness a true double-edged sword. The authors conclude with a discussion of implications for both theory and practice in our increasingly connected 'small world' and suggest directions for future research. Key concepts include:
- Problem solving requires two important and complementary forms of search: searching for information (for the facts that may be important pieces of the puzzle) and searching for solutions (for theories that combine puzzle pieces into an answer).
- The same network structure can promote or inhibit knowledge diversity, depending on whether that knowledge consists of information, or interpretations of information.
- 'Good' communication structures may only be good for parts of the process of collective problem solving: structures that are good now may be bad later.
- Organizations might be wise to adopt different communications structures for different phases of collective problem solving.
- Rather than allow the march of technology to dictate organizational performance, it is possible to imagine how technology could be harnessed to achieve different performance goals.
Using data from a novel laboratory experiment on complex problem solving in which we varied the network structure of 16-person organizations, we investigate how an organization's network structure shapes performance in problem-solving tasks. Problem solving, we argue, involves both search for information and search for solutions. Our results show that the effect of network structure is opposite for these two important and complementary forms of search. Dense clustering encourages members of a network to generate more diverse information but discourages them from generating diverse theories: in the language of March (1991), clustering promotes exploration in information space but decreases exploration in solution space. Previous research, generally focusing on only one of those two spaces at a time, has produced inconsistent conclusions about the value of network clustering. By adopting an experimental platform on which information was measured separately from solutions, we were able to reconcile past contradictions and clarify the effects of network clustering on problem-solving performance. The finding both provides a sharper tool for structuring organizations for knowledge work and reveals the challenges inherent in manipulating network structure to enhance performance, as the communication structure that helps one antecedent of successful problem solving may harm the other.