AN INSTRUMENT FOR THE DEVELOPMENT OF THE ENTERPRISE ARCHITECTURE PRACTICE

Abstract

In this paper we introduce an architecture maturity model for the domain of enterprise architecture. The model differs from other existing models in that it departs from the standard 5-level approach. It distinguishes 18 factors, called key areas, which are relevant to developing an architectural practice. Each key area has its own maturity development path that is balanced against the maturity development paths of the other key areas. Two real-life case studies are presented to illustrate the use of the model. Usage of the model in these cases shows that the model delivers recognizable results, that the results can be traced back to the basic approach to architecture taken by the organizations investigated and that the key areas chosen bear relevance to the architectural practice of the organizations. 1 MATURITY IN ENTERPRISE ARCHITECTURE Increasing complexity in the overall information systems portfolio of an organization, and especially in the integration of information systems, requires enterprise architecture as a guiding principle. For this to work, sound architectural practices have to be implemented. Enterprise architecture, however, is a relatively young field (Bucher et al, 2006), (Lankhorst et al, 2005). Architectural practices still have to be established. There appears to be a need for an instrument to support and accelerate this. In this paper we will introduce such an instrument. We will start with distinguishing three basic types of architecture maturity models: the staged 5-level models, the continous 5-level models and the focus area oriented models. Enterprise architecture being a relatively young discipline, we feel that at the moment most organizations benefit best from the focus area oriented model. We will therefore introduce an architecture maturity matrix that falls into this category. 1.1 Architecture Maturity Models Most maturity models are concerned with software development and maintenance. The most widely used is CMM and all its variants (CMMI, 2002). Recently some architecture maturity models have been developed. These models are all based on the generic 5-level maturity model used by CMM. Two variants can be distinguished. 1. Staged 5-level models. These models distinguish five levels of maturity. For each level a number of focus areas are defined specific to that level. These focus areas have to be implemented satisfactorily for the organization to achieve that particular level. 2. Continuous 5-level models. These models also distinguish five general maturity levels and a number of focus areas. The difference with the first kind of models is that the focus areas are not attributed to a level, but within each focus area the 5 levels are distinguished. Searching for models that were not build around the standard five maturity levels, we also looked at other process maturity models and found a third type of model from test process improvement (Koomen and Pol, 1999): 3. Focus area oriented models. These models depart from the idea that there are five generic maturity levels. Instead each focus area has its own number of specific maturity levels. The overall maturity of an organization is expressed as a combination of the maturity levels of these focus areas. The differences between the types of models is illustrated in figure 1. Figure 1: Three kinds of architecture maturity models. Looking at the three types of model, we prefer the focus area oriented model because it allows a more fine-grained approach, making it more suitable to our purpose of developing and improving the architectural practice, rather than merely assessing its current maturity: The focus area oriented model makes it possible to distinguish more than five overall stages of maturity. This results in smaller steps between the stages, providing more detailed guidance to setting priorities in developing the architectural practice. Departing from the five fixed maturity levels makes the focus area oriented model more flexible in defining both focus areas and interdependencies between focus areas. In our opinion this better fits the current state of maturity of the architectural practice, where complex combinations of many different factors determine its success. The application of the first two kinds of models to architectural processes can be found in the literature in various forms. The US Government Accountability Office uses a staged model (GAO, 2003). Examples of the continuous model can be found in (Appel, 2000), (METAgroup, 2001), (NASCIO, 2003) and (Westbrock, 2004). The continuous model may also be found as foundation for various kinds of organization readiness assessments like for instance the Net Readiness Scorecard that measures the preparedness of an organization to make use of the internet-based economy (Hartman et al, 2000). Application of the third kind of model to architectural practice we have not encountered yet. This is why we decided to develop a model based on the third type ourselves, the architecture maturity matrix. Another approach to organizational improvement is the balanced scorecard approach (Kaplan and Norton, 1992). The balanced scorecard is used to evaluate corporate performance, not only on financial aspects but also on customer perspective, internal processes and learning capability. The balanced scorecard concept has also been applied to the IS function (Martinsons et al, 1999). The main difference between the balanced scorecard approach and the model presented in this paper is that the balanced scorecard is concerned with setting specific perfomance goals, while our approach is concerned with how to reach such goals.