Component Business Model

I first heard about IBM's Component Business Modeling CBM approach in a speech by Sam Palmisano (speech, my original comment). At first, it seemed as if this indicated a move by IBM into the subject of my 2001 book: Component-Based Business (book, blogpost).

I have now had a chance to look at some examples of CBM in practice.

• Component business modeling: A private banking example (June 2004)
• Component business modeling - mapping the way in insurance (June 2004)
• Component business modeling in retail banking (June 2004)
• Financial services firms prepare for an on demand world (July 2004)

As far as I can see, IBM's method is pretty close to the Information Engineering methodology, which I practised in the 1980s. The business component model (or component business map) looks remarkably like an old-fashioned business system architecture or functional decomposition. It is drawn as a neat stack of rectangular boxes, arranged into three management levels (direct, control, execute), without any explicit interaction between the boxes.

In the materials I've looked at, IBM doesn't tell us how it arrives at this model. The model is closed, internally focused, and apparently set in concrete. There seems to be little attention to the coupling between the business components, or to the interfaces between the business and the outside world, which might seem to be important architectural concerns. The case studies seem to be largely at the ISP (information strategy planning) level, rather than business strategy.

This kind of model supports a divide-and-conquer agenda, and is undoubtedly useful for managing the IS portfolio, as the IBM case studies indicate. However, it doesn't deal with some of the system-wide emergent properties and structural complexities of the large enterprise, into which my own component-based business approach was an early venture.

For more on the component-based business, see my Architecture blog.