Information Systems Research Vol. 23, No. 2, June 2012, pp. 376–396 ISSN 1047-7047 (print)  ISSN 1526-5536 (online) http://dx.doi.org/10.1287/isre.1110.0373 © 2012 INFORMS Institutional Contradictions and Loose Coupling: Postimplementation of NASA’s Enterprise Information System Nicholas Berente Terry College of Business, University of Georgia, Athens, Georgia 30602, berente@uga.edu Youngjin Yoo Fox School of Business, Temple University, Philadelphia, Pennsylvania 19122, youngjin.yoo@temple.edu T hrough a grounded analysis of the National Aeronautics and Space Administration (NASA’s) enterprise information system (IS) implementation in the months immediately following the go-live, we show how NASA can be characterized as an institutionally plural organization, rife with diverse institutional logics, some consistent and some contradictory to each other. The enterprise system is introduced in accordance with the logic of managerial rationalism, but some of the institutional logics that organizational actors draw upon and reproduce contradict the logic of managerial rationalism in certain situations. In these situations, organizational actors loosely couple elements of their practices from the practices implied by the enterprise system, thus satisfying the demands associated with both institutional fields. We identify four generalizable forms of loose coupling that result from these institutional contradictions: temporal, material, procedural, and interpretive, and discuss their effects on both the system implementation and local practices. Further, we show how, through the use of institutional logics, researchers can identify fundamental institutional contradictions that explain regularities in the situated responses to enterprise system implementations—regularities that are consistently identified in the literature across a variety of organizational contexts. Key words : institutional pluralism; institutional contradiction; institutional logic; institutional theory; loose coupling; loosely coupled; enterprise systems; ERP; NASA History : Michael Myers, Senior Editor; Natalia Levina, Associate Editor. This paper was received on August 6, 2008, and was with the authors 19 months for 3 revisions. Published online in Articles in Advance July 26, 2011. A loosely coupled system is not a flawed system. It is a social and cognitive solution to constant environ- mental change  Loose coupling is to social systems as compartmentalization is to individuals, a means to achieve cognitive economy and a little peace. —Weick (2001, p. 44) 1. Introduction If we are to take decades of social studies of enter- prise system-related practices seriously, we might conclude that enterprise resource planning (ERP) sys- tems cannot possibly be successfully employed across diverse organizational contexts (Pollock et al. 2007). Such systems are premised on notions of standard- ization, integration, and inscribing stable, generic best practices across entire organizations; but no two organizations are identical, their environments are constantly changing, and they are comprised of dis- similar groups with an array of potentially conflicting goals and interests. These diverse communities are notorious for idiosyncratically appropriating enter- prise systems—partially, improvisationally, inconsis- tently, or unfaithfully (e.g., Ciborra 2000, Scott and Wagner 2003, Wagner and Newell 2004, Pollock and Cornford 2004, Boudreau and Robey 2005)— inevitably undermining the very premise of standard- ization and integration on which the implementation of enterprise systems are based. Thus we see a number of failures in the literature (e.g., Lee and Myers 2004, Barker and Frolick 2003, Al-Mashari and Al-Mudimigh 2003). However, organizational vic- tories with ERP are also evident in much of the literature. ERP systems have been widely adopted across a variety of organizations, and ERP-related expenditures represent perhaps the single most signif- icant portion of modern corporate information tech- nology (IT) expenses (e.g., Davenport 1998, Weill et al. 2002, Sambamurthy et al. 2003, Volkoff et al. 2005, Gattiker and Goodhue 2005, Srivardhana and Pawlowski 2007). How is it, then, that ERP systems can often be successful in practice, while their promise is so obviously unachievable? 376