Enterprise Architecture (EA) is a discipline that addresses how the elements of an organization fit together, today and in the future, and how these elements transition to support the organization’s strategic plans. Not only is EA a useful tool for technical developers, who can design more consistent and interoperable systems and solutions, but it also supports managerial decision-making and helps align strategy with the structure. An advanced EA also comes with analytical methods that enable various kinds of impact analyses on hypothetical change scenarios, facilitate capital planning, and help sequence IT development.
In this paper, we propound that architectural work in an enterprise be designed and built around organizational accountability levels and be divided into three distinct yet interlinked architectures: Technical Architecture, Socio-Technical Architecture, and Ecosystemic Architecture. Each of these architectures would be self-contained and self-regulated with its paradigmatic function, methods, and tools. This would promote the principle of “separation of concerns” and also help assigning the ownership and stewardship of different architectural artifacts to governance roles that are rooted in the respective organizational levels.
Technical Architecture has an operational focus and is geared to present-day value realization. IT planning is a rational, deterministic and economic process that aims at reliability, operational efficiency and IT cost reduction. This is the realm of traditional IT architecture, information systems design and development, enterprise integration and solution architecture work. It also addresses architectural work practices and quality standards, e.g. architectural support of implementation projects, development guidelines, and change management practices. Governance is focused on compliance, internal control and risk management.
Cause and effect relationships are relatively immediate and largely visible. Structured techniques and processes ensure repeatability and predictability. The decision model is to sense incoming data, categorize the data, and then respond in accordance with predetermined practice. The behavior of systems is treated as a “black box”. Operations are optimized with one-dimensional economic indicators, while the internal design is not afforded high importance.
Socio-Technical Architecture plays an important role as the link between strategy and execution. The business strategy is translated to the design of work and the organization. Architecture is about creating enterprise flexibility and capability to change rather than operational optimization: the focus on reliability is balanced with focus on validity in anticipation of changes, whose exact nature cannot be accurately predicted. Governance addresses not only conformance but also performance and highlights strategic considerations such as value creation and resource utilization.
Cause and effect relationships are separated over time and space, which calls for expert know-how, systems thinking and analytical methods. The decision model is to sense incoming data, analyze the data, and then respond in accordance with expert advice or interpretation of that analysis. Human judgment and analytical support for decision-making grow in importance. When viewing the organization as an open and adaptive socio-technical system, knowledge about its internal operation and construction is of essence. The organization is seen as a “white box". The purpose is to design the enterprise coherently so that enterprise strategy may be executed utilizing all its facets, including IT.
In Ecosystemic Architecture, the enterprise is designed systemically vis-à-vis its environment, to enable co-evolution with its business ecosystem, industry, markets, and the larger society. In order to survive, the organization must transform in sync with the ever-changing ecosystem. This calls for organizational innovation and sustainability, wherein business increasingly follows IT. Governance is about coordinating inter-organizational forms such as business networks, alliances and public-private partnerships around a shared purpose.
Due to the interdependencies and nonlinear interactions between organizations in the ecosystem, the patterns of cause and effect are discernible only in retrospect. The decision model is to create probes to elicit the patterns, then sense those patterns and respond by stabilizing the desirable patterns, while destabilizing the undesired ones. Traditional architecture methods, tools and techniques render inadequate, while e.g. narrative techniques are powerful, as they convey a large amount of knowledge or information in a very succinct way. The organization is seen from outside of the proverbial box, as a co-evolutionary constituent within the broader business ecosystem.
The discipline of Enterprise Architecture (EA) is still relatively immature and incoherent. The discourse is rather fragmented and lacking a shared vocabulary. To shed some light on the situation, some schools of thought on EA have been suggested, each with its distinct concerns and set of assumptions. In this article, we aim to bring more structure and clarity to EA discourse. Not only do we review the identified types and schools of EA, but we also attempt to make sense of the underlying structural and metaphysical underpinnings of the field and to ground EA in theory. As per our analysis, requisite architecture methods and tools are contingent on the level of complexity. In particular, while best practices and linear techniques are applicable in a contained operational scope, they fall severely short in addressing complex problems pertaining to non-linear discontinuities inherent in the increasingly interconnected and global business environment. On the other hand, we view that an ideal scope of an architecture “work system” is bounded by a maximum number of people able to create a shared meaning. Accordingly, we propose that architectural work in an enterprise be divided into three distinct yet interlinked architectures: Technical, Socio-Technical, and Ecosystemic. Each of these architectures is self- regulated, based on different ontological and epistemological assumptions, has its own vertical scope, and requires its own distinct methods and tools.
