Tuesday, September 4, 2007

People a key factor in EA.

Many of the challenges in EA relate to people i.e. where people don't have the right motivations, capabilities, personality and background. One needs to assume of course that they are neither misologists nor misoneists (though sometimes this is hard to assume based on the evidence and the attachments to tools/mechanisms that they have long used without any obvious success - see http://ea-in-anz.blogspot.com/2007/09/ea-cant-be-done-with-document-or-case.html; http://ea-in-anz.blogspot.com/2007/08/uml-is-not-language-suited-to-most.html).

The comments below on capability and personality type come from research I have recently read.

Motivation
In many cases people don't have the motivation to see EA done properly
  • self-interest: if my knowledge represents my value then why should I share it? And if everyone can share knowledge without me - why do they need me.
  • insecurity: if my judgements are based partially on gut instinct (to be charitable) or blind prejudice (to be uncharitable) I don't want my judgements to be examined critically i.e. the basis on which I reach my conclusions (develop my plans and strategies).
  • secrecy: frequently people don't want to disclose their plans or strategies because it allows people to oppose them. Usually within the organisation this is counter productive.
  • laziness: thinking is hard, takes time and effort, and if I can wing it - then that may be easier.
  • mypoia: I am judged based on this years projects/results (not how well it works when I am gone).
Capability or way of thinking
EA's should:
  • be visionary and be able to conceptualise
  • be able to analyse and problem solve
  • show business acumen and be aware of situational politics
  • Most importantly be able to communicate - and perhaps not annoy with their constellations of abilities ;-)
It seems that IQ is a key determinant, along with communication skills, business knowledge, problem solving and attitude towards time (i.e. re the future). Not age, intuition, technical skills.

Personality type
EA's should be:
creative, open minded, passionate,engaging and resilient

Background (knowledge, skills and training)
It is useful to have had a broad background i.e. technical (hardware, software, services, applications etc.) and business (sales, marketing, operations) and to have had training in design (engineers, architects etc.) and communications.
To have developed this breadth of background takes time.

To often an EA is criticized by some point technologist for not understanding the specific domain of the point technologist well enough. This is common criticism of architects generally.

The Peter Principle and EA.
When an organisation doesn't have hard and concrete measures of function (e.g. EA) the "Peter Principle" often comes into play.

Often the EA will be well respected within the Enterprise in an abstract kind of way (i.e. people will say "he knows a great deal about our business and systems", "he is very smart", "he knows lots about technology") , but not in a concrete way (i.e. "we really know what he does, what he produces etc."). This usually presents an EA who has risen through the ranks of technologists to pop out at the top. They will think that EA is little more than an extension of the technology oriented role they most recently had (e.g. SW design, infrastructure design/CMDB, Business process design etc.).

Frequently these people are ill-suited to EA based on their cognitive/personality profiles and they don't have broad backgrounds (i.e. they are essentially technologists - interested in technology).

Conclusion

To be success in EA one needs someone with the right: motivation, capabilites, personality type and background. These people are hard to find. The capabilities and personalities are very hard to change - but it is comparatively easy to provide people training to supplement gaps in their background, and it is easy to set up measures that will adjust their motivation.



5 comments:

Tom G said...

Useful info, as always.

Agree with you on the fear etc issues: it's been well documented in issues of knowledge-sharing, by others such as Dave Snowden, Shawn Callahan and others from the former IBM Cynefin team.

Will admit I'm surprised you say "IQ is the key determinant" - do you have a source for this? (For example, I would have thought EQ would be at least as important: IQ tends to be a good measure of analysis, but not the synthesis that's also needed for enterprise-architecture.)

Sally Bean's article "The Elusive Enterprise Architect" provides one of the best summaries I've seen of the broader skill-set needed for enterprise-level architecture and business-transformation - would hope you'd find it relevant here.

MJE said...

The view on IQ came from a presentation ("New competencies for EA, IT & business staff", BTELL, EA Conference, Aug 2007) by Keith Frampton. In speaking with Keith I understand his conclusions are based on field work/analysis e.g. fairly large surveys of EAs.

He quotes the "The Bell Curve" (Herrnstein & Murray). Obviously this book has generated a lot of debate.

MJE said...

I think Sally Bean's article useful (and I agree with the sentiment) but works on a flawed analogy.
I would agree that: "an enterprise has an architecture even if it doesn't have electricity"; "A virtual team is needed to meet the challenges"; "[EA is] never ending process"; "role is challenging for a number of [reasons] 1/ championing a change of behaviour that is unnatural for most people (cross-functional/ future-orientated); /2. [dealing] with a very wide range of stakeholders; /3 intelligent selection and application of tools/techniques/approaches that will work for the organisation"; "skills required include:Communicators and change agents; Visual system thinkers and modellers with foresight; Fast learners; Principled pragmatists; Incisive consultants and troubleshooters; Big picture thinkers"

I think the analogy flawed i.e. EA is more akin to town planning i.e. sets of buildings and common infrastructures than it is to architecture i.e. of an individual building. I also think a number of assumptions about the differences between architects an EA are incorrect: e.g. in neither case does the design start from scratch e.g. an architect starts with the site (slope, location, views, noise, boundaries, egress, soil/sub-soil, drainage, vegitation etc.), and its services (electrical, roading, water, drainage etc.), then there are all the codes (which codify standards of interconnection, styles/patterns of construction etc. ). These codes make clear that even an individual building is not built in isolation. Buildings will usually be built to specific support a set of functions, often an organisation, sometimes extant furnishing/equipment/plant etc. it is true that sometimes building are built to support generic functions/organisations (but these are more analogous more to OTS SW).
See: http://ea-in-anz.blogspot.com/2007/09/ea-and-analogies-with-built-environment.html

Tom G said...

Yup, The Bell Curve did indeed generate a lot of controversy. :-) The bell-curve concept is valid in itself, but could be misleading here.

"EA's should be: creative, open minded, passionate,engaging and resilient", you said. Yup, agree - and note that IQ is not only not "a key determinant" in most of these characteristics, but when over-emphasised is more like a key deterrent. For the kind of political and holistic requirements for much EA work, most of the research suggests that the 'key determinants' for those characteristics are measures such as EQ (emotional intelligence - see e.g. Wikipedia reference) and SQ (so-called 'spiritual intelligence' - see e.g. CEO Refresher reference). So whilst IQ is important, obviously, we need to keep it in balance with other attributes of self.

Tom G said...

Yup, The Bell Curve did indeed generate a lot of controversy. :-) The bell-curve concept is valid in itself, but could perhaps be misleading here?

"EA's should be: creative, open minded, passionate,engaging and resilient", you said. Yup, agree - and note that IQ is not only not "a key determinant" in most of these characteristics, but when over-emphasised to the detriment of other attributes is more like a key deterrent. For the kind of political and holistic requirements for much EA work, most of the research I've seen suggests that the 'key determinants' for those characteristics are measures such as EQ (emotional intelligence - see e.g. Wikipedia reference) and SQ (so-called 'spiritual intelligence' - see e.g. CEO Refresher reference). So whilst IQ is important, obviously, we need to keep it in balance with other attributes of self.