Learning at the Boundaries: An Action Agenda for Business Analysts Rajiv Vashist Judy McKay Peter Marshall Swinburne University of Technology Swinburne University of Technology University of Tasmania rvashist@swin.edu.au jmckay@swin.edu.au Peter.marshall@utas.edu.au Abstract The business analyst role was established in response to concerns about a ‘gap’ between technical IT staff and users. Such a ‘gap’ has been implicated in the failure of information systems to consistently deliver business value. However research suggests that while business analysts may facilitate the relationship between technical staff and business users, boundaries may still occur between business analysts and user groups on one hand, and business analysts and IT staff on the other. Nonetheless, these boundaries present opportunities for all the participant groups to learn by developing new understandings, knowledge and practices, thus enabling them to more effectively interact with their constituents. Through the adopted practice perspective, our research demonstrates that learning via socio-spatial and instrumental-developmental expansions at the boundaries can improve the practice of business analysts. Our findings make a contribution to the practice perspective literature by highlighting the importance of boundaries as opportunities for learning. 1. Introduction To address problems in information systems development (ISD), significant research efforts have been invested over the years in both the methods adopted in ISD [12] and the human factors associated with participant groups in ISD such as users [21], analysts [28], and designers [7]. While this research has contributed to our knowledge of ISD from a number of perspectives, the research on both the methods adopted in ISD and the human factors associated with participant groups in ISD tends to regard these as separable, and has typically focussed on either methodological interests, or the human interests in ISD. We argue that we may need to view the human factors that influence ISD and the methods and tools employed in ISD as being inextricably linked. Such a linkage is achieved by adopting a practice perspective [17]. Such a perspective adopts a more holistic view and emphasises how the actors, and the methods and tools employed by the actors, define a community, such as a community-of-practice (CoP) [49], and thus supports enquiry into and knowledge creation about ISD as a series of interactions amongst participant groups, with each group being viewed as a distinct CoP. For example, ISD may be viewed as a series of interactions amongst several participant groups or communities such as users, IT staff, and business analysts (BAs). Researchers have argued that many of the problems in ISD are located in the differences between these participant groups [8, 35, 46]. These differences are a source of boundaries in organisational work [50] and they become evident in the perceptions and expectations of the people in boundary interactions [14]. For example, the various participant groups in ISD are viewed as having different perceptions of reality [8], framing problems differently [46], and having different technology frames [35]. In addition, differing groups apply different logic to problem solving and possess differing objectivity orientations [41]. Despite the useful insights that arise from the research reported in these papers on how to improve practice [10], IS failure has continued to be a cause for pessimism [19] and a problem for IS researchers and practitioners [40]. To address these concerns, we need to advance these insights by finding ways to improve the work practices of diverse participant groups. In this paper we argue that if the boundaries between participant groups continue to be viewed merely as ‘gaps’, problems are likely to persist. Instead, we argue that these boundaries should be viewed as opportunities for learning, and thus for improving ISD practices involving users, IT staff, and BAs. While there is a growing interest in the notion of practices and boundaries in a number of disciplines [9, 13, 39], research specifically dedicated to understanding boundaries is limited [20]. Indeed, even research on boundary spanning is argued to have paid little attention to examining and understanding the boundaries themselves [32]. Furthermore, there is comparatively little research regarding the leveraging of knowledge at the boundaries of practices, except for some notable exceptions e.g., [6]. Interestingly, the notion of learning at boundaries has remained largely implicit [1] and thus under-researched. By investigating boundaries, we may usefully draw attention to the ways in which the boundaries create challenges in organisational work. Understanding boundaries not only provides insights for managing interactions between organisational communities but it also allows us to view boundaries as ‘spaces’ with