I want to discuss today an important point that is overlooked (or not given its due attention) in most architecture initiatives. It is a “current state architecture”. It is not uncommon that some architects jump on to developing target architecture with its associated implementation roadmap based on some requirements without current state architecture. Such architects may question the value of current state architecture because developing current state architecture may take time and incur additional cost (especially for large organizations). So, what exactly is the value of current state architecture?

The answer is simple. Assume you have stomach pain. The pain is so strong that you could not perform your daily work and you decide to go see your doctor. Your doctor asks you – what is the problem? You tell him you feel sever pain in the stomach. Now, the “target” for you and the doctor is to “feel better or relief the pain so that you get back to your normal routine”. How you get to that “target” is what matters now. If your doctor is like those architects who don’t see the value of “current state”, he will subscribe you a medication and tells you to go try that. Assume he is like that. You went off and spent your time and money – but your pain is even getting worse. You then decided to call 911 which took you to the emergency hospital. The doctor at the hospital asks you the same question – what is the problem? You tell him your problem. This doctor now tries to understand your current situation through series of examinations – blood, stool, temperature, pressure, etc… The doctor then comes back to tell you your exact current situation (current state) and tells you the steps (roadmap) he is going to take to treat the illness so that you will be well again (target state).  You get the right medication only after your current situation is understood and you are not going to spend your money and harm yourself  trying many different medications hoping that one of them will hit the target.  It is not hard to see the value of the current state architecture. Its value is much greater than the associated cost and time of developing the current state architecture.

There are two possible outcomes when one attempts to do target architecture and roadmap without current state architecture.

• Either the architecture is shelved without real implementation because of cost of unrealistic roadmap or plan resulted from lack of current state understanding, or
• The architecture implementation is done with considerable cost with less value to the business stakeholder.  Again due to the lack of current state architecture the target architecture implementation cannot correctly address the challenges or issues that needed to be solved. Moreover, it takes longer time to get to the target because of poorly understood current position.

TOGAF clearly defines the need for current state architecture – it uses the term “Baseline Architecture”. Understanding current state helps in understanding the current stakeholders concerns. Without such basic understanding, the architecture is done in a vacuum without real base.

If current state architecture has such an importance, then the next question to ask is – how much of the current state need to be done? To answer this question, it is helpful to know the scope and objectives of the architecture initiative. Any architecture initiative needs to define clear scope and objectives. Based on the scope and objectives, the current state architecture is done to the level that is enough to highlight the current issues/challenges/needs to propose target state to the stakeholders. You will clearly know what components are to be retired or replaced or changed. In addition to that you will also identify building blocks or components that are reusable in the target state architecture. By understanding stakeholders concerns by focusing on the relevant current state areas, you will be well equipped to address the various concerns in the target architecture. This makes it easy for the stakeholder to be satisfied with your vision and bless (approve) your target architecture. Your roadmap will also be realistic and expectation is properly managed.

Bye for now -:)

Awel Dico