Architecture Is Not Optional

Once again I’ve been asked: “but do we really need an enterprise architecture”? And as always my response is: enterprise architecture is not optional, and a better question would be: “do we really need to need to manage our enterprise architecture”?

Enterprise architecture is not optional – it exists regardless of whether we choose to make it explicit or manage it formally. Enterprise architecture results from the ways in which we combine, structure and use a vast array of IT and business building blocks. But it can evolve in a haphazard way, without any formal architectural roadmap or strategy. In fact, for many enterprises this is exactly how the architecture does evolve.

So the question is: what advantages or benefits result from attempting to manage this evolution? And the answer is that the difference between unmanaged and managed, or between managed well and managed badly, can be quickly and easily measured and demonstrated!

The best measures are along the lines of the following examples:

  • How many different software or hardware products provide roughly the same functionality? Would it be better if there were fewer? What would be the savings if you didn’t have the overheads (maintenance and updating, training, multiple business processes) of so many software and hardware products?
  • How many separate projects across the enterprise are making similar changes? Would it be better if these were coordinated so that they could share resources and costs?
  • How often do changes impact other areas in unexpected ways? Would it be useful to proactively and easily assess the consequences of change ahead of time?
  • How often is the wheel reinvented? Would it be useful to build a core set of building blocks that could be used over and over again, rather than building things from scratch? And would it help if those building blocks represented the core strengths and uniqueness of the enterprise – making it difficult for competitors to copy?
  • How often is change a reaction to outside pressures? Would it be better to outsmart competitors and do something different, rather than copy market leaders?

These questions lead to interesting discussions that can make a big difference to the future of the enterprise. They are questions that can only be answered when the enterprise architecture is managed.

And so once again, if you are ever asked: “but do we really need an enterprise architecture”?, the response should always be that enterprise architecture is not optional, and that a better question is: “do we really need to need to manage our enterprise architecture”?