If you’re not familiar with Todd Biske’s blog then do yourself a favour and have a read. Todd is pragmatic and business focussed. And he’s also an Enterprise Architect. A rare combination. I normally agree with Todd but his last two posts here and here got me thinking. These posts talk about Centres of Excellence and whether they should be temporary or permanent. My question is: why do we need them at all?
Todd quotes a meeting of the SOA Consortium which talked about the skills required to operate a SOA Centre of Excellence. They are: project/portfolio management, service design, business knowledge, technical aptitude, communication, teaching and governance. Todd also mentions Gartner’s idea of an ‘Integration Competency Centre’.
As far as I’m concerned ‘SOA’ and ‘integration’ are subsets of enterprise architecture. Furthermore, all of the skills listed above should be found in your existing Enterprise Architecture Group. So why do we need more centres, groups, committees or offices? Todd’s answer:
While enterprise architecture has the appropriate scope of visibility and influence on the technical side, they’re not the best group for handling organizational decisions.
This is a damning indictment of the EA profession. Worse than that: it’s true.
That said, I think there are three main reasons why creating a new group might be a bad idea.
How is this new team going to be staffed? I’ll assume for the moment that getting business people will be easy (big assumption). But what about the technical people? Your best guys will probably be in the existing EA Group. In most organisations SOA is not the only game in town. Who is going to look after the other things when the best architects are poached by another group? If the EA Group is large it will typically have lots of specialists – so who will you take and who will you leave behind (take the information architect and leave the infrastructure architect)? And how will you fill the resulting hole? If the EA Group is small it will usually consist of generalists but taking even one person from a small group will leave a significant hole in its ability to deliver.
2. Overlapping Responsibilities.
There will be huge overlap between the functions of the new group and the existing EA Group. When responsibility is shared, accountability suffers. No ONE is accountable for anything. How will you divide the responsibilities between the two groups? How ill you manage the inevitable overlap (however small)? Will the business have to go over everything twice?
3. Cross-functional teams are usually temporary.
Cross functional teams are nothing new. But typically these groups are short-term and temporary – formed for a specific purpose (project) and then dissolved. For example, businesses may use these teams to implement new quality systems, oversee a merger or implement a new CRM system. The simple reason cross-functional groups are temporary is because the best people are reluctant to get off their career path within the business and move into a ‘special project’ permanently. SOA is not a ‘project’ and it’s certainly not ‘short term’.
A Better Solution?
If your EA Group can’t deliver on everything that a Centre of Excellence would be expected to – then fix your EA Group! Typically the two main problems are a lack of credibility and a lack of business focus. Both of these can be solved by making the EA Group more business focussed. This can be done in two ways:
Short term solution. Physically bring business people into the EA Group. You’ll then have a cross-functional centre of excellence without actually creating one from scratch. Of course, there won’t be too many business people volunteering for the job, but this is not an insurmountable problem. A properly functioning EA group has enormous potential for delivering business value – and the business person responsible for ‘finally getting those techies to see things from a business perspective’ will be a popular fellow at the executive golf day. So there’s plenty of upside for any business person wishing to take on the challenge. And it should only be a temporary assignment.
Long term solution. To do their jobs properly EA’s MUST build relationships with all parts of the business. If the business avoids engaging with the EA team (or vice versa) then there’s a major problem. If this isn’t fixed then you might as well dismantle the EA group as it would be impossible for them to be adding value to the business.
In my opinion, creating a new group to overcome the shortcomings of an existing one is akin to saying “my house is dirty so I’ll go buy a clean one” (thanks Krusty of Simpsons fame for this quote). The new house will get dirty eventually. Similarly any new group created will eventually face and same challenges that the EA Group did (the day to day ‘stuff’), and probably end up in the same position.