Inside Microsoft we have a twice yearly set of technical training for the people in the field called Tech Ready. I have just signed up to do a session and was looking through the list of other session to see what else was being covered. I noticed that there were quite a lot of Enterprise Architecture sessions and that got me to thinking that I should really blog about EA, especially as I was the first person (in fact still am the only person) to write a EA whitepaper for Microsoft back in 2002. At that time I was convinced that EA was the only way to address the complexity and integration issues that existed within the Enterprise. I thought that if you created a present "as is" model of the enterprise which had the correct level of granularity and then built a "to be" desired state model then you could run a series of projects to get from one to the other.
Well 5 years on and I am older and maybe a bit wiser. I talked to lots of really smart Enterprise Architects and a common concern was what to do with these architectures when they were built. How did you stop them becoming "shelf ware" of limited or no value? How could you justify the whole of the Enterprise Architecture profession? I, along with a lot of other people, were convinced if you had the right starting and ending model and a rigorous, governed process then all would be well. I saw a lot of really good models but things didn’t seem to be working out quite the way I expected. Even though people were building great models they were not producing anything of any value to the organization. At first, along with everyone else, I blamed the lack of good process and Governance and Enterprise Program Management became the crie de jour. Enterprise Architects world wide began to focus on and implement good process models but things still did not seem to change much, EA was still shelfware and EA still had trouble justifying itself. About 18 months ago I became uneasy that there was a fundamental flaw in the whole concept of EA and stopped talking about it, indeed I have since been recommending that MS doesn't provide any more guidance around EA.
Working with the Web 2.0 in the enterprise stuff I have begun to realize what the problem with EA is. It's the "to be" model that is an issue; most "to be" models are just a clean architectural implementations of what the organization needs today they are not what the organizations need for the future. If the whole of EA is just a cleanup exercise then it's hardly of any great benefit or interest to the business side of the enterprise and in fact is just yet another hard to justify cost reduction exercise. The business side of organizations is pretty fed up with paying year on year to clean up the mess that IT has got itself into and EA is seen as just another program to do that, just like SOA. The business now want value creation and real revenue not "productivity" and perceived cost savings.
So how do we create business value from Architecture? How can be use technology to give business a real competitive advantage? How can be use models to generate revenue? These are not questions that EA can answer. We need to go back to the issues that business has and look at how we can use architectural techniques to solve them which is what Strategic Architecture is all about.
Architecture to me is all about the use of abstraction and models (in the widest sense) to manage complexity (by simplification), allow improved communication and provide the ability to forecast. These principles can be used to design a program, manage a server farm, ensure a system is scalable, create new business processes or even create new business value.
Architecture can be used to create new business value by examining the complexities of the business domain, the marketplace or social domain and the technical or technological domain. From this we can abstract out the important elements, simplify them so they can be communicated to other, in particular business people, and then forecast what is going to happen in the future that will create opportunity for the business. This strategic architecture is different from other architectures insofar as it majors on forecasting rather than complexity simplification. It's output is new opportunity areas and new projects for a business to build.
Strategic architecture is the same as application or infrastructure architecture insofar as it has a set of standard models (Business, Social and Technical), patterns within those models and even processes to undertake a strategic architecture study (see Hagel's work). It is the most important architecture within a business because it creates future marketplaces, value and revenue for the organization. It is not Enterprise Architecture, it goes far beyond enterprise architecture in scope and value, and yet is by far the least well understood and documented of all the architectures.
As organizations begin to better understand the value of architecture for complexity management, communication and forecasting we will see the dramatic growth of strategic architecture and the creation of the strategic architect role. This will become the most valuable and valued architectural role in an organization. Enterprise Architecture will be seen to be an interesting but irrelevant turn in the growth of architecture, the future will be built around a Strategic Architecture Methodology (SAM).