In a recent Webinar on the role of enterprise architecture as enabler for business decisions the following statement was made: “Current enterprise architecture models are too simplistic to handle complex and dynamic business problems“. The solution was a systemic modeling of the situation and interrelations, especially for enterprise architecture. Generally this is an interesting approach, but ultimately it is just a model for the perception of reality – an inaccurate picture of reality. Therefore it has to be judged as differentiated as every model. It depends on its use and the target.
What really had scared me was the conclusion. I have had the impression that the complexity of the models is already very high. All possible kinds of situations have to be illustratable and realizable in all kinds of environments. Therefore these models are mostly just metamodels. The meaning of complexity is that you cannot forecast the outcomes. This is the case for all of these models; there is no way of predicting the outcome or the quality of the outcome. The complexity of a system is growing with the number of elements, links between these elements and the functionality of the links. No certification can help; it is inherent in the system. So today’s models are already complex.
Let’s deconstruct the sentence step by step, to make clear what the main issue is:
Current Enterprise Architecture Models : mature frameworks like TOGAF 9 etc.
are too simplistic : these models oversimplify reality
to handle complex : this simplified image of reality makes it impossible to handle complex situations (but is this not the idea of simplification?)
and dynamic business problems: the impact of changing influences on the business have to be evaluated quickly.
What we see is that the main problems are not models but dynamic influences. These influences have a huge effect on the existing solutions in the user landscape and therefore also on the business. The more exactly you create a model the lower is your speed when analyzing impacts on the business.
Unfortunately hardly any company was able to identify the impact of smartphones and tablets in time – no matter how mature their EA organization or frameworks has been. Today, the early identification of chances is more relevant than an exhaustive image of the situation. What you need is fast reactions when making a decision, when planning a measure as well as concerning your application landscape. The system architecture also has a huge influence on the latter. After all, the only thing that counts is: time to market.
