EA is not dead … Long live the EA

[Latest News Headlines]

In his latest blog post - Is Enterprise Architecture dead - Richard Veryard argues the demise of the role of EA stating:

"I can't see much recognition or respect for the value of "Enterprise Architecture" except among the ranks of EA practitioners. In many organizations, the function of enterprise architecture is squeezed or marginalized, if not rejected altogether."

A related blog post from David Sprott on The Death of Enterprise Architecture also suggests that EA's are less than successful and the profession is in a state of decline.

To be honest, its not the first time I've heard these arguments, both in terms of the role as well as the function. There seems to be more discussion on the pitfalls of EA today (see related Gartner article blog post) as opposed to the benefits EA can bring. Is the fundamental issue that the value/scope of the role/practice is simply misunderstood by both business and IT and that organizations are simply hiring the wrong people into the role ? Do EA's have the right sponsors within their organizations ?

Lets start with the Profession. A subsequent post by Veryard on the EA Profession states that it suffers from associations with Professional Organizations (e.g. The Open Group) that have relatively few members and that have fundamentally failed to communicate the business value of architecture to the industry at large. Furthermore their certification methods/criteria are based on theory/one-time examination rather than real world practice and little is done in the way of maintaining quality assurance and governing against maverick or incompetent practitioners. Whilst I have to agree with these sentiments, I don't agree that you can judge a profession on the basis of its associations or governance structure. Professional bodies such as the Open Group are formed to provide guidance, operating frameworks, methodologies and principles. Organizations looking to implement the profession often use these as professional blocks rather than, for example, implementing the entire TOGAF methodology and framework. Furthermore, large multi-nationals such as HSBC are firmly committed to professional bodies such as the Open Group (all their Solution/Enterprise Architects have been or are in the process of obtaining TOGAF 9 certification) and additionally most of my customers are asking for an Oracle mapping to TOGAF.

This of course does not prove that there isn't an issue with the EA role or profession. I have read other Blogs that argue that the role is required but its the EA reporting line that is the real issue - i.e. the EA role should report into the COO as opposed to the CIO as it provides the relevant level of business/IT neutrality to the role whereas typically the EA reports into CIO/CTO where the business may perceive the architect to be more IT than Business focused. Moreover, it's rare I consistently meet the same type of EA. They come from a range of backgrounds ranging from business to IT and they are all in different stages of maturity in terms of their practice of EA. And here we get to the fundamental issue, whilst EA is not a new profession, neither is it a mature one. Can you truly describe an immature profession as being dead or at best in a state of decline ? Based on the same criteria, couldn't  you describe all other Architect professions (e.g. Business, Solution, Information, Infrastructure etc....) as being less than an overwhelming success ?

In my humble opinion, the profession, job title or reporting line is not the issue. EA is firmly grounded on a set of principles/skills/behaviours associated with helping to achieve valuable/measurable Business-IT Linkage. People that exhibit those behaviours should have a firm understanding of both business and IT to be successful in their role and to be honest, there aren't many of these guys out there hence the problem and why I agree to some extent with Veryard and Sprott's blog comments.

Secondly, the issue often lies in the word 'Enterprise' which implies the EA might only operate at the 'Enterprise' level, never dipping their toes into business domain areas/projects, primarily focusing on the 'To be' rather than helping to resolve the 'As Is' hence never gaining the credibility to work at the 'Enterprise' level. In my experience, EA's are now focusing more on 'quick win' or strategic company initiatives rather than architecting the enterprise. They have seen the light in the sense that you have to gain credibility in practical 'bite size' chunks (i.e. aimed implicitly at optimizing business processes in stages) and use those smaller (project/program) wins to enhance your reputation in the business. The golden rule here though is to ensure you are not compromising the most effective Enterprise 'To Be' vision through your work/focus on smaller initiatives - i.e. always keep the Enterprise view in mind.

Finally, to some extent, the word 'Enterprise' may by some be interpreted as a level of seniority over other disciplines such as Information and Solution Architecture when in fact all of these disciplines are dependent on one another in defining and implementing/governing the ideal future state vision.

Let's not start to get excited about the demise of EA. The Business will always require IT to meet its business goals and they will always require a visionary in IT that would help them to succeed in the most efficient use of IT.  However, the following principles need to be applied to get EA back into the 'Circle of trust':

1. Getting the right person exhibit the right skills/behaviours/competencies in the role
2. Second your EA's to the business - more than often, the problem with EA's that they do not sufficient understand the business they work in to give them the credibility to architect for the future. At the recent Open Group Conference in London, the then CIO of Transport for London (TfL) waxed lyrical about the lack of business understanding in the EA team and confirmed he was in the process of seconding his EA's into business facing roles.
3. Ensuring the optimum reporting line to ensure optimum business buy in - it might not be the CIO
4. Ensuring the EA is sufficiently aware of the day to day operational issues and is seen to be suggesting resolutions in those areas (as a set of tactical measures to gain business credibility) as opposed to working in a 'Future state' vacuum
5. Ensuring there is a mechanism that track the metrics associated with the 'value' of EA that is understood and accepted by the business
6. The EA functions works closely with other architectural disciplines - e.g. Information Architects. These professions should be treated as inter-dependent rather than independent. Ensure that the optimum business & IT governance processes are in place so that the professions works closely and effectively together as opposed to against each other
7. Always consider the outcome the business is looking for as opposed to what EA deliverables you think the business might need - all too often, the EA function is delivering shelf ware as opposed deliverables of real business value. Spend time talking to the business about what they expect from you.

Ok, its not that simple to implement these principles but they can help to bring the EA back from the dead. We at Oracle still see the value of investing our time in effort in providing thought leadership to the EA community and will continue to do so for the foreseeable future.

EA is not yet dead, long live the EA !



More from the Enterprise Architecture at Oracle Blog ...