Performance Tradeoffs in Team Knowledge Sourcing
Bradley R. Staats
University of North Carolina Kenan-Flagler Business School
Harvard Business School
Amy C. Edmondson
Harvard University - Technology & Operations Management Unit
October 14, 2011
Harvard Business School Technology & Operations Mgt. Unit Working Paper No. 11-031
This research examines how teams organize knowledge sourcing (obtaining access to others’ knowledge or expertise) and investigates the performance trade-offs involved in two approaches to knowledge sourcing in teams. One approach a team can take is to specialize, such that a small number of members source knowledge on behalf of the team. This specialized knowledge-sourcing approach lowers search costs. The other approach has most or all team members engaging in knowledge sourcing. This broad approach means that more team members interact directly with the knowledge source, and thus may understand the knowledge better. These options present a sourcing paradox: teams cannot reap the advantages of specialized sourcing and the advantages of broad sourcing. They face performance tradeoffs. Further under some conditions performance tradeoffs will be more pronounced. Specifically, specialized knowledge sourcing depends on within team knowledge sharing, and so conditions that hinder knowledge sharing in a team are likely to reduce the effectiveness of the specialized approach. Using archival data from several hundred software development projects in an Indian software services firm, we find support for most of our hypotheses. Our findings offer insight for theory and practice into how team organization, organizational knowledge resources, and within-team knowledge sharing can aid team performance.
Number of Pages in PDF File: 35
Keywords: Knowledge Management, Knowledge Sharing, Knowledge Sourcing, Learning, Performance Tradeoffs, Teamsworking papers series
Date posted: September 18, 2010 ; Last revised: September 22, 2012
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo8 in 0.500 seconds