This study addresses a major gap in our knowledge about the allocation of information technology (IT) decision rights between business and IT units at the application level, including the governance of applications delivered on-premise versus those delivered with a software-as-a-service (SaaS) model. Building on the findings from a multicase qualitative study of organizations that had adopted the same SaaS application, we draw on three theoretical lenses (agency theory, transaction cost economics, and knowledge-based view) to develop a theoretically grounded model with three organization-level factors, three application-level factors, and application-level IT governance. Hypotheses derived from the model, as well as a set of differential hypotheses about factor influences due to on-premise versus SaaS delivery, are tested with survey responses from 207 firms in which application-level governance is operationalized with two dimensions: decision control rights (decision authority) and decision management rights (task responsibility). Three antecedents (origin of the application initiative, scope of application use, business knowledge of the IT unit) were significantly associated with application governance postimplementation, and the on-premise/SaaS subgroup analyses provide preliminary evidence for the mode of application delivery as a moderator of these relationships. Overall, this study contributes to a growing body of research that takes a more modular approach to studying IT governance and provides theoretical explanations for differing application-level governance designs.
The management of the information systems (IS) function is a complex task, particularly in the case of multinational corporations (MNCs), where installations dispersed across distance, time, and cultures can lead to diverse and incompatible systems spreading among foreign subsidiaries. The need to globally control and coordinate the IS management function is often met with resistance from local IS managers, who may perceive corporate standards as intrusive. Resource dependence theory (RDT) argues that control is made easier when a subsidiary unit is dependent on corporate headquarters for critical resources. This study examined the IS management relationship and the use of various mechanisms of control (formal and informal) between 54 headquarters--subsidiary pairs spread across 19 countries of varying resource-richness. While RDT appears to be valid when subsidiaries are dependent on MNC headquarters for resources, the expected relationship between the mechanisms and host country IS resource availability was not observed. Although there was a significant relationship with the use of informal mechanisms and IS resources, it was in the opposite direction to what would be expected by RDT.
This study contributes to the growing body of literature on the value of enterprise resource planning (ERP)investments at the firm level. Using an organization integration lens that takes into account investments in complementary resources as well as an options thinking logic about the value of an ERP platform, we argue that not all ERP purchases have the same potential impact at the firm level due to ERP project decisions made at the time of purchase. Based on a sample of 116 investment announcements in United Stateś based firms between 1997 and 2001, we find support for our hypotheses that ERP projects with greater functional scope (two or more value-chain modules) or greater physical scope (multiple sites) result in positive, higher shareholder returns. Furthermore, the highest increases in returns (3.29%) are found for ERP purchases with greater functional scope and greater physical scope; negative returns are found for projects with lesser functional scope and lesser physical scope. These findings provide empirical support for prior theory about the organizational integration benefits of ERP systems, the contribution of complementary resource investments to the business value of IT investments, and the growth options associated with IT platform investments. The article concludes with implications of our firm-level findings for this first wave of enterprise systems.
Increasingly, scholars and practitioners acknowledge that information technology (IT) human capital is a strategic resource and that its effective management represents a significant organizational capability. We use configurational theory to examine organizational practices related to the management of IT human capital. In contrast to much prior work in IT human resource management (HRM) that is focused at the individual level, our inquiry is focused at the organizational level of analysis. Building on strategic human resource management (SHRM) research in general and research on the management of IT professionals in particular, we examine the broad question: Are different configurations of IT HRM practices associated with different IT staff turnover rates? A multidimensional view of IT HRM practices is presented, based on prior IT and SHRM literature. We formalize hypotheses regarding the relationship of turnover with configurations of IT HRM practices grounded in prior theory and empirical research. Based on survey responses from 106 organizations, IT HRM dimensions and configurations are derived and the hypotheses are tested. A five-configuration solution, obtained via cluster analysis, includes two contrasting configurations consistent with two archetypes found in the prior literature. Specifically, the configuration with a human capital focus has lower turnover than the task-focused configuration, providing support for our first hypothesis. Although the hypothesis on intermediate configurations and their relationship with turnover is not supported, we discover and interpret three additional configurations that embody patterns of practices with unique emphases. Theoretical and practical implications of the findings are discussed.
Horizontal mechanisms are structural overlays (such as roles and groups) and non-structural devices (such as physical colocation) that are designed to facilitate cross-unit collaboration. The objective of this study is to increase our cumulative knowledge about what top-down mechanisms are being used to promote the coordination of IS activities across corporate/division boundaries. Propositions about how mechanism usage differs under centralized versus federal IS organization contexts are developed based on a synthesis of prior organization science and empirical IS literature. Multiple methods are used to collect data from IS and non-IS senior managers from two case sites with theoretically different IS coordination needs. As predicted, multiple types of structural and non-structural mechanisms were implemented for business-IS coordination in the company with a centralized IS context, and for corporate IS-decentralized IS coordination in the company with a federal IS context. An unexpected finding was that mechanisms for both of these kinds of IS coordination were valued at each case site. The prediction that a formal group mechanism would be perceived as more effective for achieving cross-unit coordination than an integrator role mechanism was not supported. The article concludes with a discussion of implications for research and practice.
The prior IS literature points to the importance of organizational context for predicting a firm's IS governance solution. However, for the most part this literature assumes that firms adopt a uniform IS governance solution for all business units and that this solution can be predicted by context variables at the overall organization level. The purpose of this study is to increase our knowledge about why firms implement a hybrid IS governance solution in which a subset of IS functions that includes systems development is decentralized to some business units, but not to other business units, in the same enterprise. A theoretical framework of context variables at the business unit level is first developed. An embedded, single case study provides an initial test of eight propositions derived from the framework, as well as an opportunity for theory building. Data are collected utilizing both deductive and inductive methods from IS and non-IS executives of a divisionalized Fortune 500 firm in which a uniform decentralized solution for systems development in place for almost a decade has recently been replaced by a hybrid solution. The case study findings suggest that a configuration of four variables characterizes a business unit context conducive to decentralized systems development governance (organic decision-making, high business unit autonomy, a differentiation competitive strategy, and an unstable industry environment). As predicted, however, the influence of these variables is likely to be overridden and a "deviant" solution adopted when deficiencies in IT capabilities are perceived and there is a culture that supports structural change at the business unit level. Additional interview and survey data collected from the key stakeholders are then analyzed in order to develop a richer understanding of the dimensions of the IT capabilities construct at the business unit level. The notion of absorptive capacity provides a theoretical argument for the emergent findings. Implications for researching today's increasingly complex IS governance forms are then drawn.
Change-readiness is the ability of an information systems (IS) organization to deliver strategic IT applications within short development cycle times by utilizing a highly skilled internal IS workforce. This paper examines two important questions: What are the design elements of a change-ready IS organization? How can transformations to such designs be effectively managed? Insights to these questions are generated through a case study of the conceptualization and implementation of an innovative organization design within a large IS unit at Bell Atlantic, a Regional Bell Operating Company. First, a rich description is provided of the components (strategy, structure, processes, people skills, reward systems) of a centers of excellence (CoE) design that has yielded measurable gains in IS performance. Then, an analysis of the two-year implementation of this CoE design is provided in terms of anticipated and unanticipated change actions, as well as a summary of "lessons learned." The conclusion describes the design as a model worthy of consideration by other IS managers for developing change-readiness IT capabilities. Comparisons with other models for the IS organization are then drawn.
Identifying the best way to organize the IS functions within an interprise has been a critical IS management issue since the mid-1980s. Yet to date, MIS researchers have offered little empirical evidence on which to base guidelines for the practitioner. This study seeks to explain a firm's IS organization design decision for a decentralized, centralized, or "hybrid" locus of responsibility from an expanded set of environmental, overall organizational, and IS-specific antecedents as well as a larger concept of organizational alignment. Potential antecedents (drivers or enablers) are selected from prior contingency research and the IS literature; other variables emerge from the data collection. Data collected via on-site interviews from IS and general managers in six multi-divisional firms, paired by industry, confirm that centralized, decentralized, and hybrid IS structures exist-but often not in "pure" form-and that industry type is not a strong predictor. Data was also collected via survey form to capture ratings of importance for drivers (for enablers) for a recent IS design change in each firm. Based on both qualitative and quantitative data, four configurations are discussed: patterns of antecedents that are associated with (1) highly centralized or (2) highly decentralized IS structures; and patterns of antecedents that explain a firm's choice to (3) decentralize or (4) recentralize systems development and application planning functions in particular. A model based on these configurations is then proposed. The article concludes with implications for researchers and practitioners.