I confess to my own naivete. Perhaps it’s from all those years working at the level of the “Enterprise Architecture Profession” and not getting into the gory details of any single line of business. Perhaps it’s from the difficulty in creating models out of the blue, and feeling like “why would anyone want to do that twice?” Maybe it’s just from placing too much emphasis on writers like Porter and his models of business, and not enough time working with the leaders of different businesses. Maturity happens in real time.

Up until recently, if you would have asked me, “should there be a generalized business capability model that industry-specific models should derive from?” I would have answered “yes”.

I was wrong.

There have been industry specific capability models for decades, of course. TM-Forum created one of the best early models and for years they had no competition. Many other industry groups from banking to supply chain started out with process models and stumbled toward creating a capability model. But most industry groups have caught up.

However, these are not “centralized” models. They break down the core concepts of business in a way that uses a metaphor that is comfortable and familiar to the executives in that line of business. High tech would break down their capabilities around product management, engineering, sales, and operations, for example. The supply chain business would break down their top level capabilities around “plan, source, make, and deliver”. Oil and Gas uses top level capabilities focused on exploration, exploitation, transportation, refinement, and distribution. These are fundamentally different metaphors that simply make sense for their business models.

Can centralized models be created? Sure. There are some great efforts that have gone into creating a single central model and deriving industry specific models from it. That can be a good place to start for those industries that are so fragmented, or informationally immature, that there is no “industry council” seeking to set standards. But for those industries that do have education, standards, and an effort at maturity, starting with industry level metaphors, instead of a centralized model, is simply a better approach.

Many authors have talked about the notion of “MECE”. We often hear that a good capability model is MECE: Mutually Exclusive, Collectively Exhaustive. In other words, every capability in the organization is in the capability model. No one is left out. And just as importantly, every capability is in the model exists once and only once. This attribute is key to understanding how software and technology can be used to meet business needs.

I guess I’ve finally come to terms with a fact that many of my colleagues may have accepted a long time ago (and perhaps others haven’t really thought about yet): As long as the model is MECE, it does not matter what the metaphors are.

In other words, as long as we have all the bits, and we have them in the model once and only once, it doesn’t matter how those bits are organized.

Will there ever be a single model that stands as the starting point for all other models to call “mother”? Probably not. And there probably shouldn’t be.

Before I got married I had six theories about bringing up children. Now I have six children — and no theories. (unknown author)