Website Design

(Repost) WHY is EA broken?

(While on holidays, I thought I’d repost some of my more popular articles. Here’s one from October 19 that had ~2400 views. – Neil)

On Friday, I posted an interesting article from Jason Bloomberg entitled “Is Enterprise Architecture Completely Broken?” that was published in Forbes. I thought it was a really well written article (no pandering here, Jason) and thought it accurately portrayed the current state of Enterprise Architecture. It asks questions that I’ve posted before – what does it mean to architect an Enterprise? What do we need to do to fix the practice of EA?

And the article finishes with the statement –

“Let’s put down the red stapler, stick that framework in a filing cabinet, and focus on practical steps to achieving the long-term transformation goals of your organization” 

But here’s the thing that we have to remember – we are Architects and, as a result, to come to a solution we need to first understand the current state. To create a strategy on how to get to the TO BE state, we have to have the AS IS situation understood.

We have to apply those things we learned in creating solutions in an Enterprise in creating a solution to a broken EA.

So, let’s start with the current state of EA. And let’s be honest here. By and large, I see a lot of pontification about the importance of EA. I see talk about frameworks and I see Architects talking about various details about EA. But, when I enter a client’s environment, what I actually see is a completely different story.

I see EAs that focus on big projects and not strategies. I see EAs that may create strategies but those strategies aren’t manditory in being followed. And I see EAs that don’t talk to Lines of Business outside of IT. In short, I don’t see EAs.

I see big picture Solution Architects.

Which tells me that, in a SWOT analysis, the strengths are associated with the Technical expertise (hence the reason why you see EAs all assigned to specific TECHNICAL domains like Application, Infrastructure, Security, etc.). But that also tells me about the weaknesses. There isn’t any communication going on with the Lines of Business to understand the direction of of the business itself. And that’s most likely because most people will stay with where they are comfortable – in this case, EAs are most comfortable with Technology.

I’ve written a number of blogs including one dealing with a question someone asked me titled “What does an Enterprise Architect do?”. The comments that I’ve received to all the blogs are wide ranging. I’ll paraphrase a comment that I’ve received since it’s applicable to this topic;

“If we can’t agree to what we do amongst ourselves, how can we explain it to the CXO?” 

Okay, that’s fine to say looking at the EAs themselves. But what about the Organizations. Remember, when you create strategies, you have to look at the Environmental Variables that the Strategies have to take into consideration. So let’s do that in this case:

  • Variable 1 – The Executive layer doesn’t think in technical terms, they think in business terms. (Social variable)
  • Variable 2 – EA tends to be placed in the IT departments, so are viewed as technical in nature by the Executive layer. (Social & Technical variable)
  • Variable 3 – PMO is responsible for the delivery of Projects, the “visual” part of strategies. (Political & Competitive variable)
  • Variable 4 – Budgets are controlled outside of EA. (Economic variable)
  • Variable 5 – The larger an Enterprise gets, the more difficult it gets for communication to flow (to and from EAs). (Social variable)
  • Variable 6 – Enterprise’s, while talking about being process oriented, move away from set processes the larger they get. (Social variable)
  • Variable 7 – People want what they want, when they want it (Social, Technical, & Competitive variables)

I could go on with quite a few other variables but I think this will suffice for now. The core issues I see associated with those variables are that

a). communication is poor to and from EAs and

b). relevance is tied to results.

From a communication point of view, I think that EAs are too focused on the technical and not enough on the business. And the people that matter in terms of determining the relevance of Enterprise Architecture is the business, not the technical folk. So, for EA to be relevant, I would suggest that EAs should probably be removed from the IT environments and put into the business in some way, shape or form. Closer to the business and further away from the technical. EAs need to start talking to business IN THEIR LANGUAGE so that engagement is easier for them.

They are our customers!

From a relevance point of view, you have to look at how we show results for the Executive suite. Are solutions being delivered in a timely manner? I would suggest that, until results are shown by EAs, they will continue to be viewed more and more as irrelevant. Solution Architects – sure. And that’s easy considering they work on projects and projects can show concrete results. But EA? Much harder.

The Executive layer is required to show results and typically over a shorter term that EA strategies. So, if the Executive layer is expected to deliver quicker than EAs, then how can EAs help them if they focus on the longer term? Maybe it’s like someone else mentioned in another of my blogs. To paraphase;

“The CEO is the only true Enterprise Architect” 

So, let’s summarize;

EAs don’t communicate well with the business. That’s got to change.
EAs don’t provide results in the time frames that the Executives and the business need to provide results. That has to change.
That last one, btw, is the bigger issue. And if we aren’t providing support in the timeframes that the business and, more importantly, the Executives need, then we will be slowly viewed as more and more irrelevant. Frameworks were meant to speed up the process. But they become rather burdensome because people keep adding layers to them all in the name of “improving” results. There was a Software Architect that commented in one of my blogs something to the effect of “Agile doesn’t work in frameworks” (again, I’m paraphrasing so sorry if I didn’t get that comment correct).

Maybe there’s a need for an “Agile” approach to EA? Seems counter-intuitive since EA looks longer term.

Maybe the issue isn’t so much EA as much as it’s Program delivery of strategy and a communication associated with activities.

Let’s make sure we first truly understand what’s causing the problem in the first place. Then we can provide solutions to fixing ourselves.