Mark McDonald

A member of the Gartner Blog Network

Mark P. McDonald
GVP EXP
8 years at Gartner
24 years IT industry

Mark McDonald, Ph.D., is a former group vice president and head of research in Gartner Executive Programs. He is the co-author of The Social Organization with Anthony Bradley. Read Full Bio

Coverage Areas:

Enterprise architects, creationism and absolutism

by Mark P. McDonald  |  November 19, 2012  |  15 Comments

This blog post is going to get me into  trouble, but I think that the issue is worth discussing, so I wince as I hit the publish button, but here goes.

Recently I had the opportunity to listen to an Enterprise Architect describe enterprise architecture, its importance and role in an organization.  It was an illuminating discussion.

The enterprise architect who spoke saw architects as master designer of the enterprise, the person who knows everything about the enterprise, its systems, processes, organization and relationships.  In the presenter’s own words

Enterprise architects compose holistic solutions that address the business challenges of the enterprise and support the governance needed to implement them. 

That got me thinking about creationism.  Creationism sits in opposition to the theory of evolution advocating that the world was created by a single all knowing, all seeing, all doing deity who composed and ordered the world.  The creator as intelligent designer is responsible for the intricacies, developments and mutations normally associated with evolution. The enterprise architect speaking likened himself and the discipline of enterprise architecture to being that ‘intelligent designer.’

According to the Wikipedia definition, an enterprise architect is a “person responsible for performing this complex analysis of business structure and processes and is often called upon to draw conclusions from the information collected.”  Continuing from Wikipedia, “as the purpose of architecture is: “INSIGHT, TO DECIDE, FOR ALL STAKEHOLDERS”, enterprise architects work very closely with the enterprise sponsor and key stakeholders, internal and external to the enterprise. The architect understands the enterprise mission, vision and strategy and the sponsor’s ideas about the approach.”

If this is not the description of at least a demi-god, then I do not know what it is.  I am not trying to poke fun at EA’s.  I respect the work they do and I have played that role myself on more than one occasion.  But the degree to which EA has puffed itself up requires letting some air out of the role and making some room for the rest of us in determining the future of our enterprise.

They and EA in general is falling into a trap of absolutism.  The idea that there is only one way, one authority, one direction for something as complex as a business.  It is the kind of trap that Nicholas Taleb covered in his excellent article in this weekend’s Wall Street Journal “Learning to Love Volatility.”  He points out that rigid structures, the kind that form around absolutes are vulnerable to volatility.  Too often these structures are ‘architected’ by people who believed they knew better.

Are enterprise architects the ‘creators’ or ‘intelligent designers’ of their organizations?  It certainly seems so in the definitions and descriptions used by many enterprise architects, who believe that without architecture an enterprise cannot exist.  They see ‘architecture’ as bringing structure along absolute principles and guidelines.  Is the Zachman framework all that different from Aristotle’s classification scheme, or the four bodily humors that guided medicine for more than 1000 years?  From Zachman’s own site, italics added by me.

“Since The Zachman Framework™ classification was observed empirically in the structure of the descriptive representations (the architecture) of buildings, airplanes and other complex industrial products, there is substantial evidence to establish that The Zachman Framework™ is the fundamental structure for Enterprise Architecture and thereby yields the total set of descriptive representations relevant for describing an Enterprise.”

Enterprise architects start on a slippery slope based on the explicit comparison of an enterprise with other man-made entities, such as airplanes, buildings, and other industrial products.  EA’s conclusion that a company is fundamentally the same as other engineered products means that they believe an organization can be analyzed, designed, and governed in the same way. That is the fallacy gives EA a  ‘creationist’ flavor, unfortunately in  human and social organizations there are few absolutes.

The issue here is that EA’s  see themselves as the unique and ultimate intelligent designer of the enterprise and woe to an organization that operates differently. Enterprise architects profess a supernatural understanding and power over the enterprise. The discipline believes that it brings order out of chaos, strategy to the masses, tell us our place (organization), our work (process), our tools (technology) and how our value (metrics).

Enterprises are not man-made artifacts that can be deconstructed, analyzed, designed, and optimized like machines or buildings. Enterprises are social constructs that evolve organically around explicit and implicit agreements between the diverse group of stakeholders and actors.

An enterprise is a social-technical entity driven by humans.  Enterprises are not entirely organic or always self-emerging but a curious mix between design and emergence that is hard to describe from a purely logical/analytical perspective.  The idea that enterprise architects are the only ones with the breadth of knowledge, intellectual horsepower and perspective to design how an enterprise works is a business equivalent to divine intervention.

Enterprise architecture needs a reset

I point this out because I believe that enterprise architecture and many enterprise architects have gone too far, become too full of themselves and too full in their belief that architecture trumps strategy, the CEO, the market, etc.   The result is that what should be a highly effective management tool is compromised into academic debates of the need for and what constitutes enterprise architecture.

Now I know that not every EA or architecture supports this extreme, but after hearing this enterprise architect speak I just had to ask myself – who  does he think his is and whom does he think he is talking to?

It is a sign of weak analysis to believe that any one person, one team or group can place itself above all others in the modern corporation.  It is also a sign of weak management to endorse such a view that leads to strategic constipation when we need strategic participation. Such extreme hierarchicalism and hubris are clear signs of How the Mighty Fall according to Jim Collins and others.

This is one of the reasons why I believe that the future of IT rests in an enlightened approach to technology rather than a renaissance of classic IT practices – that includes an enlightened approach to enterprise architecture.

A word of explanation before we open the floor to comments

Before you comment on this post about they author’s intellectual capacity, understanding of architecture, appreciation for the challenges facing organizations, etc.  Please be aware that I believe that architecture is an important tool, that an overall view is important and we need to manage complexity.  So this is not against EA as a discipline or practice, but rather that it’s worth considering that it’s gone too far and needs a reset.

So let the discussion begin.

15 Comments »

Category: EA Personal Observation Technology Tools     Tags: , , , , , ,

15 responses so far ↓

  • 1 Lee Johns   November 19, 2012 at 2:29 pm

    You can find me as @StorageOlogist on twitter so this seemed like an appropriate post to respond to.

    Enterprise architects have a tough job because of the previous enterprise architect who left them with a legacy of equipment that is not best suited for the needs of today. Alot of their energy is spent designing fixes for things that would not need fixing if you started with a blank sheet of paper.

    I totally agree that their needs to be an enlightened view across the board. I see to many folks who can not accepet a better way simply because their belief in what they deal with everyday prevents then from realizing that new alternatives exist.

  • 2 Larry Reid   November 19, 2012 at 9:37 pm

    An excellent, thought provoking piece. IT has to evolve and diversify, not standardize and stagnate. Today’s “perfect solution” (as if such a thing existed) is tomorrow’s “legacy of equipment that is not best suited for the needs of today.”

    An enterprise architect is valuable if she understands the business, and takes a very hard line on providing true business value through IT. Merely parroting vendors’ claims of value is not enterprise architecture.

  • 3 Kai Schlüter   November 20, 2012 at 1:07 am

    Thank you for sharing your view. I think this is a great blog post. My own observation is that there is quite often a confusion between the function Enterprise Architecture and the role Enterprise Architect, especially if someone is titled EA.

    I personally see Enterprise Architecture indeed as the holistic function, but I do not believe that this should be carried out content wise by the Enterprise Architects alone. Instead Enterprise Architects should help the organization to come together in a structured approach (EA function) and helping to seek for the answers.

    I think this is similar to the Project Management function, where also a lot of people act as Project Manager according to the PM function, even though they do not have an official PM role.

  • 4 Elizabeth Stocks   November 20, 2012 at 4:44 am

    Interesting article and yes a bit controversial!
    I think you need to start with the concept that Enterprise Architecture is about Strategy and Business.
    Unfortunately EA seems to have got stuck in IT or is run out of the IT department; hence we now have to take an extreme view of EA to get it back on the business agenda.
    Like any movement there have to be extremists who feel passionate about the place of EA and that is what the speaker was doing, pushing the boundaries of people’s perceptions.
    EA is there to interpret the business strategy and provide strategy roadmaps (motorways, A-roads and B-roads). As most companies use IT to support the business capabilities which are in the strategic vision, this has compounded the confusion as to where EA lies.

  • 5 Elizabeth Stocks   November 20, 2012 at 5:45 am

    Try reading this article, as the Chief Strategy Officer might be the answer.

    http://sloanreview.mit.edu/the-magazine/2012-fall/54110/the-role-of-the-chief-strategy-officer/#.UKc_yy3dgWY.twitter

  • 6 Adrian Grigoriu   November 20, 2012 at 6:46 am

    While it raises the blood pressure a tad, your opinion is right. It is true that “the idea that enterprise architects are the only ones with the breadth of knowledge, intellectual horsepower and perspective to design how an enterprise works is a business equivalent to divine intervention.”

    Is it just an honest mistake? I think it’s rather hypocritical self advertising in a chaotic market that has little criteria of evaluation for the EA end product and the frameworks that would help deliver it. It works though.

    Enterprises have been born today without an EA. EA though, even documented after the fact, would provide an edge to the enterprise in this utterly competitive world. At least because it improves the understanding and provides a common vocabulary and plan for the enterprise.

    At the end of the day, EA still is the blueprint of the enterprise. It can and should be equally used by all stakeholders (CxO, strategy, operations, programmes, architects. sourcing…) to document the enterprise, fix its operation, improve, roadmap it, implement its strategy and govern it. That is still mighty impressive if it is executed properly which is mostly not.

    The main task of the EA is to deliver that blueprint for the stakeholders to be able to achieve the above outcomes, unless a job description specifically states differently.

    The reality is also that today EA happens and serves the IT alone. The dream is that the EA would serve the whole enterprise. But the these two are often confounded.

    It is odd though that this post comes from the organisation which, for what it matters, is not far from stating the same thing, that EA is “leading enterprise responses to disruptive forces…”.
    See this quote from Gartner’s glossary:
    “Enterprise architecture (EA) is a discipline for proactively and holistically leading enterprise responses to disruptive forces by identifying and analyzing the execution of change toward desired business vision and outcomes. EA delivers value by presenting business and IT leaders with signature-ready recommendations for adjusting policies and projects to achieve target business outcomes that capitalize on relevant business disruptions…”

    Check this blog out: http://it.toolbox.com/blogs/ea-matters/enterprise-architecture-state-and-causes-iii-53236
    It is a series of posts on the EA state.

  • 7 Mark P. McDonald   November 20, 2012 at 7:10 am

    I want to thank everyone for their comments to date and provide some responses.

    To Adrian above, you are right that Gartner as a firm advocates a particular definition and type of EA that is supporting the view I describe in the post. That is part of the reason I mentioned that this post will generate some issues. It is not that I think that Gartner’s definition is wrong, incorrect or ill advised, I just wanted to point out a trend, implications of where EA appears to be headed and offer an opinion.

    I believe accurately defining something requires debate and discussion to move things forward. The EA descriptions out there, some of which are quoted in the post, represent a rather strong and polar definition of EA based on an assumption that one can absolutely know and design the enterprise. I was just trying to point out the implications and limitation of this view. Taleb’s article in the Wall Street Journal really put a head on this idea.

    To Elizabeth, Adrian, Larry and Lee — yes the “enterprise” in most definitions of EA is largely centered around IT. I do believe this is a limitation, but stating EA’s role as designer of the entire enterprise is not an answer for lifting EA out of its IT centricity. Also I believe that existing EA frameworks remain IT centric, hierarchical, decompositions of things capturing the stock of resources and their classification than the flow and interaction of resources to create value.

    Thank you all for your comments and thoughts as well for understanding the purpose of this post is to open a discussion about where EA has defined itself, the implications of that definition and its practicality in a world that is increasingly dynamic, disruptive and volatile.

  • 8 Tom Graves   November 20, 2012 at 10:43 am

    Hi Mark – This is a really useful post, and I’m delighted to see this problem being properly addressed by you here: several of us had a not-quite-shouting-match recently with folks from another major consultancy that got it wrong in exactly the way you warn about above.

    Kai hits the nail right in his comment above, about the distinction between enterprise-architecture as a business-need, versus enterprise-architects who enact (part of) the work to fulfill that need. Chris Potts likewise gets it right in his book ‘recrEAtion’, where he argues that the actual ‘chief architect’ for the organisation is the CEO: some of that work may be delegated to those with ‘enterprise-architect’ in their job-title, but also to strategists, investment-managers, portfolio-managers and many others.

    To me it’s not that enterprise-architects ‘design the enterprise’, but more that their work would focus on facilitating the co-creation of that ‘enterprise-design’, bottom-up as much as top-down, and outside-in as much as inside-out. It’s difficult to describe in precise terms, because it’s different with every type of organisation, every industry, and the maturity and context of the individual organisation and of its respective enterprise-architecture.

    One point we do know is that EA as a discipline should not be placed solely as a sub-unit of IT: doing so causes all manner of problems and failure-risks, especially over the longer term. It’s often useful to start EA in IT, but not wise to keep it there. For more on this, perhaps see my post ‘Two enterprise-architectures’ http://weblog.tetradian.com/2012/10/17/two-enterprise-architectures/

  • 9 Russell Plummer   November 20, 2012 at 11:58 am

    Mark,

    An excellent post, but I wonder if one of the problems lies with the people around the EA? The EA takes & presents a holistic view, as per the comments already made.

    But if that holistic view is presented well, then the people who percieve, and more importantly not understand, that view may have a tendency to take it as the real thing rather than what it represents (“reficiation”: taking a theeoretical concept and treating it as concrete).

    In such a case I can see that the EA could be seen as delivering the whole, rather than a representation of the whole.

    Looking at your closing comments about the increasing volatility of our wider environment, then making any mistake of treating the holistic view as a reality becomes even more dangerous.

  • 10 Tom Catalini   November 26, 2012 at 10:43 am

    Mark,

    Thanks for mixing it up here, and for noting (appropriately) that management has a responsibility to avoid this extreme EA viewpoint.

    Too often, many IT related strategies are delegated rather than embraced as challenges at the highest level. Maybe because it’s hard work and a heavy responsibility (with ample risk), or maybe because the nature of the challenge is fundamentally misunderstood (prehaps it’s both). Either way, the conversation needs to be had. Thanks for bringing it to light in a thought-provoking way.

    Tom

  • 11 Jim Stikeleather   November 28, 2012 at 11:39 am

    Mark – spot on! Enterprise Architects should be encourage to pursue at least a limited study of Complex Adaptive Systems theory. It turns out that the most complex behavior, the inherent adaptability(to new environments) and agility(to respond to new situations) of biological systems are based upon just a few rules.

    Future IT systems are going to be even more complex; in that they are dynamic networks of interactions, and their relationships are not aggregations of the individual static entities. They must be adaptive; in that the individual and collective behavior mutate and self-organize corresponding to the change-provoking micro-event or collection of events. Trying to freeze them into an all encompassing operationally static (you won’t have time to re-look at the architecture every time you have to deal with change) architecture is doomed.

  • 12 David MacKenzie   November 28, 2012 at 12:59 pm

    Mark:

    My takeaway from your post is that you feel EA’s have developed a dangerously over-inflated opinion of their importance, considering themselves to be the equivalent of the supposed “Intelligent Designer” advocated by Creationists. You’ve based this on your interpretations of comments in a presentation, followed by research on Wikipedia and other blog posts. In reading up on your citations, I can’t say that I would draw the same conclusion. But you did, and others may also.

    This situation provides a perfect opportunity to explain how we Enterprise Architects view ourselves and understand our jobs.

    EA’s do not create enterprise vision, mission and strategy. Those things already exist. Without EA, though, they exist as fragmented pieces in many forms: Ideas in people’s heads, an ancient statement on a web site, annual statements to Wall Street, etc. This situation leads to a lack of unified direction across the enterprise, and at worst, differing understanding of the vision, mission and strategy. The results become apparent in technology sprawl, projects that consume resources and produce little or no business value, low staff morale and high turnover.

    The EA’s job is to aggregate all of this information and translate it into a form that removes ambiguity and provides a common understanding. This involves listening to the people who have the ideas and leading the people who do the work in a direction that realizes those great ideas. It also involves verification and validation by the original information sources.

    You are correct in thinking that EA’s feel this is a very important role. But, we understand that it is one of many very important roles in an effective enterprise. We feel that good EA job performance is critical to enterprise success. But, that can be said of any leadership function in the enterprise.

    You may be getting the impression you have based on the fact that there are many web sites, blog posts and presentations on the practice of EA. This is not the case for CEO’s, CFO’s, etc. That, however, is due to the relative immaturity of the practice and the lack of common understanding about the role.

    As a matter of fact, an EA who believes as you think he/she does is doomed to failure.

  • 13 What is it like to work as an enterprise architect? | The Career Advisor   December 2, 2012 at 12:45 pm

    [...] Enterprise architects, creationism and absolutism [...]

  • 14 Nick Malik   December 4, 2012 at 8:41 am

    Mark – I, too, have heard some of my peers make inflated statements when speaking with other architects.

    Sometimes, but not often, they make the mistake of making those statements to stakeholders. So let’s seperate the bluster that happens when speaking with peers from the more careful words that happen when speaking with actual architectural stakeholders.

    Secondly, the section in Wikipedia that you cited is an egregious example of “wiki-nonsense” that was added by someone who had not signed in to wikipedia before making the edit. I have not stayed on top of the edits to the page over the last six months, and this really odd edit crept in. I have removed it. It is nonsense.

    Before we call for a reset of Enterprise Architecture, ask the actual EA stakeholders (not the EAs themselves) to send you the descriptions of EA that they have sitting around in their inboxes and SharePoint sites. You will find a very different description.

    Respectfully,
    — Nick Malik

  • 15 PLURES ex UNIS – an architecture model for digitalized business   May 9, 2013 at 7:02 am

    [...] Enterprise Architects, creationism and absolutism [...]