I’ve been working in the same place for the past 10 years, where we’ve grown organically over two decades without much overarching strategic guidance to direct our growth. While things function alright, there are signs that our operation has become unwieldy. Quite simply, the technology that runs our operation is complex enough so that we have a hard time answering questions about which parts affect one another. There’s no question that we are large and complex enough to warrant the application of enterprise practices, but we haven’t taken it on in an organized fashion until now. Seeing the need to become more mature in the way that technology is used, a decision was made to introduce the practice of Enterprise Architecture. Enthused about helping to be part of a positive change, I made the pitch to fill the role of Enterprise Architect and lead the transformation. So, here I am figuring out how to do just that.

Ultimately, my work involves figuring out how to increase our IT maturity — Enterprise Architecture is the tool that I’ve been provided with. Just as any tool can be used in multiple ways, Enterprise Architecture can be wielded to produce many different effects. Taking this analogy one step further, if EA is a hammer, I’m trying to figure out how to drive the nail instead of smash my thumb or miss the head of the nail and damage the material.

Since all of this is quite new to the organization, I have the advantage of starting with a blank canvas. But, with great power (and potential) comes great responsibility. Going the wrong direction will just waste time and money on producing a body of knowledge about our enterprise that isn’t useful. Being pragmatic in nature, I decided very early on that whatever I do, I must make any introduction of EA useful. It’s such a simple requirement – almost too simple to state. Obviously, EA needs to be useful. But this is where the quest begins. What does it take for EA to be useful?

Cynics and Apologists

In perusing articles posted about instances of EA being implemented, I very quickly discovered skeptics declaring the failure, and even the death of Enterprise Architecture. Surprisingly, I found it common for articles that aim to put EA in a positive light, like this press release about the successful implementation of EA, to make defensive opening comments, such as “The success of using EA in large change and innovation projects is often debated”.
Naturally, I’m concerned. If the discipline of EA is a landmine, I have a sincere interest in making sure that I don’t cause it to explode. It seems apparent to me that in 2018 the EA party is winding down and the discipline has failed to reliably deliver on the promise of strategic, digital transformation. However, I still believe that it can be done right and deliver value. I don’t think the party is over yet, I just think that the right music isn’t playing.

Enterprise Architecture is a meta-discipline, and therefore, inherently dispensable

Meta disciplines are beneficial in the way that they enhance understanding of something that already exists. That is all.

The physical world exists perfectly fine without the discipline of metaphysics. Data can be produced and used in the absence of good quality metadata. Our lives carry on whether or not we apply practices of mindfulness (another current meta-discipline). IT architecture can exist and function whether there is an Enterprise Architecture used to govern design and implementation. In fact, it is most common for EA to be added only after an enterprise already uses IT to solve problems, not before.

However, as the complexity and volume of information about anything grows, so does the need to build some kind of model around it so it can be understood. Things can exist without our understanding of them. It happens everywhere in life. But if we need to keep things under control, we MUST understand how it works. For example, if we want to use data to make serious decisions, we need to understand where it came from. Is the data old? Is the data from a reliable source? Is the data complete? This knowledge of meta data will help us understand how to appropriately use the data to make decisions. As such, the understanding of how IT architecture works for the business must be present in order to make good decisions about design that will help the enterprise accomplish what it has a mission to accomplish.

Given the fact that technology solutions exist, and even may flourish, without a standard practice of Enterprise Architecture to guide them, I have a question.

What does Enterprise Architecture need to be valuable? 3 things.

The promise of EA when it is done well is extremely compelling. Especially with all of the current buzz about digital transformation that seems to declare the same promise of enabling business strategy through technology as EA does, there’s got to be value. In principle, it is wasteful to run a business enabled by technology without the practice of EA to guide the design and plan the transitional projects. What follows are my opinions about the top 3 principles that make it possible to succeed with EA.

1. Know the stakeholders
   

The first rule in any project is to know who will be affected by the project. I think the biggest mistake that can be made in implementing Enterprise Architecture is to use it as a tool for the C-Suite, and ignore the technologists, or even worse when reversing that balance. EA is about the confluence of business and technology. When the interests of one party are served without considering the interests of others, the formula doesn’t compute. Technologists need to understand how the technical bits are related, but without understanding how the business functions and strategies are supported the risk of introducing unnecessary technology is imminent. Likewise, when the business side of the architecture is documented with the supporting technology but isn’t shared with those who are responsible for implementing solutions, the power of bringing strategy to product design is lost.

Herein lies one of the inherent weaknesses of EA. It only works well when the business-thinkers design along side the technologists. It is not common for these two camps to speak the same language.

1. Keep it simple
   

There’s nothing like a massive framework and lengthy, complex processes to cause people to get lost. In EA, these kinds of frameworks abound. I find it interesting that the vendors of EA software, which are designed to capture information about the architecture so it adheres to standard frameworks, default to a framework that follows their own design rather than the standard. It’s also common to find that those who select a standard framework need to extend, customize, and change it so it serves the needs of the enterprise better. To me, this is a sign that established frameworks are too large, brittle, and complex to really serve the purposes of businesses very well. This emphasizes the point that selecting a framework and customizing it is definitely the right way to implement EA. I believe that a minimalist approach is best so that time isn’t wasted architecting for the architect’s sake. Just as the purpose of EA is to ensure that all the technology solutions serve the strategy of the enterprise, the purpose of architecting needs to trace back to decisions for which the information is absolutely required. When information about the enterprise architecture is only useful for the architect, value is delivered to no one.

1. Build it into the governance
   

In order for EA to be valuable, it needs to have a purpose. That purpose is to guide decisions about how technology can be implemented. However, just because all of the information about an architecture exists, it doesn’t mean that such information will be used to make decisions. This is where the enterprise culture and business processes need to recognize EA as an asset. When governance of the strategic implementation of technology require the use of information from the EA, there is unquestionable value. Without building EA information into the business processes of decision making, it is useless.

Is there hope for Enterprise Architecture?

Absolutely. EA just needs to keep focused on what it can deliver and how it fits into the big picture. This is the tool to use for digital transformation, but in order to be successful EA needs to serve the increasingly Agile implementation of technology by knowing who needs the information, and the simplest ways that it can be used to guide decisions.