Friday, December 31, 2010

The Independence of EA

#entarch Following my blogpost on EA and the Big Picture, @carlhaggerty asked does it matter if EA disappears into a core C-Suite competency?

Clearly it matters to some people, especially those who have committed themselves and their careers to the idea of enterprise architecture (EA) as an independent practice. If EA were to disappear, some people would be upset, and some organizations would need to find a new mission or fold.

@carlhaggerty agreed that some people might be upset but thought if it is in the best interests of the enterprise surely it would be the best thing to do.

Well perhaps. I am certainly not a defender of the current position of EA as a defacto knowledge silo, because this offends against the principles of EA itself. (See my post on A Value Proposition for Enterprise Architecture.)

But I have two problems with making a judgement about "the best thing to do" purely on the basis of "the best interests of the enterprise". The first problem is who shall speak for the best interests of the enterprise. C-level executives are often driven by extremely short-term considerations, such as stock market prices, as well as their own personal advantage. Merging enterprise architecture into C-level management might mean that EA would be forced to adopt this perspective to the exclusion of any other perspective. The second problem is that there have always been enterprises that are based on a corrupt business model - flawed or unethical - and there is clear conflict between the best interest of these enterprises and the best interests of society as a whole.

Ultimately, this comes down to the possibility of designing a complex organization to produce behaviour that is effective, intelligent and ethical. @carlhaggerty would rather see EA being done whether or not an EA exists in an enterprise, and he personally doesn't care who ;) I agree with him up to a point - I don't care who does EA, but I do want to see an organizational design giving a strong independent voice to an EA-like perspective.


enectoux said...

The main issue I foresee if EA disappears into a core C-Suite competency is that to day, no existing C-x role has the dedicated mandate to manage such EA topics.

As you mentioned, C-level are primarily focus on extremely short-term considerations, I would even add that they are finance focus (which is good up to a certain extend for the company). Having this pressure does not allow them to step back enough to get the holistic view we were speaking about on your previous post "EA and Big picture". I addition, even if I don't want to generalize it, but C-level long term generally stand for the end of the fiscal period.

So, if merging the EA into a C-suite competency means to create a new role called C-blah blah doing the same thing as a regular EA should do... What is the point?

carlhaggerty said...

The issue is not to create a new role but to mainstream the competencies that EA's have into the senior leadership of the organisation (i refer to the EA a Business Strategy) view and not the Tech Driven EA approach.

Most CEO's are actually doing EA anyway. As an EA in a public sector, my view on this will be blinkered by the sector and the cultural differences will no doubt influence my perspective.

My personal view is that I'm not doing EA as a discipline to keep myself employed, but to influence and lead change in my organisation. If by mainstreaming the Enterprise Architects competencies will actually deliver better outcomes then i'd rather see the function disappear and mainstreamed into the senior leadership within my organisation.

However this is very much a scenario and there are of course others approaches which will also deliver better outcomes. This is an evolving process and is and will be influenced by the culture.

But i stand by my view that i personally don't care who does it as long as the outcomes are being delivered.

Richard Veryard said...

Carl rightly focuses on the outcomes. My point about independence is that the outcomes may depend not only on the right activities being carried out, but on what enectoux calls "dedicated mandate" - in other words, the proper governance of these activities, based on clear responsibility and authority (separation of concerns).

(This point follows from a general point about enterprise modelling - that a purely behavioural model is insufficient to understand structure and its impact on business outcomes.)

Carl said...

Um, at the risk of sounding a little simple here, what is this Enterprise Architecture thing we are talking about? Noting that there are multiple definitions for EA and that each practitioner seems to carry their own definition in their head (exception: TOGAF folk) I am struggling to understand what is moving (or not) into the C-cloud.

Until we have a well-defined and generally agreed definition as to EA then it is unclear what the discipline is and how it needs to be developed and practiced within organisations and communities. In the absence of an agreed definition we probably have to rely on the practitioners of said black-art (which is how it feels to me sometimes :-) ) to define what needs to be done and how it integrates into the organisational psyche.

Once we know what the discipline is (and it is multi-disciplinary almost by definition) then we can position the C-cloud sponsorship for the discipline correctly and worry less about the role of “Enterprise Architect”. Until then, we need the practitioners and the wannabe’s.