Behavioral Enterprise Architecture Management

Owing to the enormous growth of investments in information systems (IS), almost all large organizations deal with hundreds to thousands of IS components. Such constant investments in IS, over many years, bring about ever-increasing IS complexity because investments often lead to new IS components and interdependencies, while existing components are barely retired. As a consequence of growing complexity, systems become heterogeneous, inconsistencies are created, and it becomes more challenging to manage IS components and their interdependencies.
Organizations typically decide on new IS investments through loosely-coupled projects, which mainly pursue local and short-term objectives. To deal with this challenge and to keep project-related IS investments in line with enterprise-wide and long-term objectives, organizations commonly employ Enterprise Architecture Management (EAM). While the potential benefits of EAM are apparent from an enterprise-wide perspective, there are often acceptance issues for project stakeholders, whose perspective is more local and short-term. Project stakeholders typically prefer solutions that create immediate local benefits, even at the expense of enterprise-wide, long-term considerations. In contrast, the benefits of EAM are only apparent in the long run and are shared throughout the organization. Therefore, many issues in EAM institutionalization may be considered as goal conflicts between long-term shared enterprise-wide goals and short-term goals of project (“local”) stakeholders. We propose a behavioral investigation of EAM to better understand IS investment decision behavior of project stakeholders in the pursuit of multiple, conflicting goals and to investigate how this behavior may be guided towards enterprise-wide goals.
Taking an institutional theory perspective, organizations can be described as complex socio-technical systems that are fundamentally described by different pressures (coercive, normative, and regulative pressures) that influence the behavior of individuals within the organization. These pressures, in the form of rules, norms, and conventions, can help to promote enterprise-wide interests by guiding the behavior of key stakeholders in individual projects. In the EAM context, the institutional theory perspective thus allows us to investigate the link between local behavior and enterprise-wide outcomes.
The proposed study would be one of the first research projects that actually goes down to the individual level of analysis and links this to the enterprise-wide level, thus taking a multilevel perspective that has been found necessary to better understand EAM phenomena (see Haki et al. 2016). To that end, the proposed research project takes an experimental approach to transfer and develop theoretical explanations from more general institutional theory studies on public goods dilemmas to the EAM context. Using experiments allows us to directly observe and analyze the behavior of individuals in controlled settings that isolate the theoretical explanation from a large number of potentially confounding factors that are specific to single organizations.
The proposal at hand presents the overall research objectives and its status quo. The project team has a reliable and internationally known track record on EAM. The involved researchers (Robert Winter, Stephan Aier, and Kazem Haki) are acknowledged senior scholars in the area of EAM, who have the required expertise and standing to develop and communicate a new approach to EAM research. In particular, the research group at the chair of Prof. Robert Winter has been at the forefront of investigating institutional theory as a theoretical perspective on EAM. Furthermore, we have conducted and published preliminary work at prestigious IS outlets (with several publications still under review), and pilot tested our experimental setup. There is now sufficient empirical evidence that calls for a thorough theoretical investigation of the combined effects of different institutional pressures on individual behavior in the EAM context, and we thus seek for financing for a concentrated research in this direction. This will help us to establish a new research stream in the EAM discipline that is expected to evidently advance the field.