Several years ago I was asked to be part of a panel discussing business process and business rules at one of the BrainStorm conference stops. The panel included James Taylor – noted thought leader and the man who coined the term “Enterprise Decision Management”. I had had the pleasure of talking with James on several occasions prior to this panel but found myself doing a double-take when he said “the dirty secret is that business analysts don’t really care about the rules”. Blasphemy! Isn’t that why we’re all here? Isn’t that what we do?
James went on to explain that while business analysts didn’t care about specific rules, they did very much care about things like “can I underwrite this loan?” or “what price should I give to this product?” These decisions were the key and this is where our focus should rightly be. I pondered this for a while and then realized that this had implicitly been our approach for a while but we’d never formalized or described it so eloquently. And there are some very good reasons to think, plan, design, implement and ultimately manage with this perspective in mind.
As we have noted previously, an organization can be most successful when looking at business rules from within the context of well-defined business processes. This core concept of Business Decision Management – looking at processes first, then the decisions that support those processes, and finally the business rules that comprise those decisions – aids the delineation of a means to reuse, manage and govern the entire scope of the business on a decision-by decision basis.
The biggest obstacle we face is that many organizations new to Business Decision Management typically think of decisions in only the most complex sense. There are, in fact, a wide range of decisions that come into play during the daily operation of any business. For example, basic data validation occurs in virtually all systems . Rules are often used to automate this functionality. This should come as no surprise. However, this type of functionality is certainly a core decision – “do I have all the data I need and is it in the correct format?” At first glance it may seem strange to think of it in this context but by doing so we have now neatly grouped a set of rules supporting this decision that may be used across many systems. We have established a context in which we can design, implement, test and manage these rules.
Decisions can obviously take on much more sophisticated forms. The kind of actions most commonly associated with business decisions are at a higher level or somewhat more complicated than the data validation example just presented. Decisions here would include things like “can I underwrite this loan?” or “how should this product be priced if purchased via a customer contract?” These decisions are key operational derivations that occur on a regular basis. Another category of decisions might not leap to mind. These include more proactive or hypothetical analyses that many organizations undertake to determine how to drive their business: “what loans should I be adding to my portfolio to mitigate risk?” or “how should I alter my pricing models to gain an edge over my primary competition?” More complex forms of reasoning in addition to rules may be involved here but we are looking at a decision nonetheless.
I hope the idea of decisions is clear on a conceptual level at this point. There’s nothing too groundbreaking here – just an acknowledgement and realization that we deal with them from a business point of view every day. So why do we care about this in more than a conceptual context? I’ve written before about the power of an integrated process / decision / rule perspective. It’s important for a variety of reasons – both technical and business. And I believe the idea of decisions is critically important as well. I’ll lay out three examples where it truly makes a difference:
DESIGN: Analyzing the rules within a business process in the context of a decision gives us a clean boundary for capture, documentation, and validation. Dealing with several thousand unbounded rules is a daunting task for anyone; handling a dozen decisions with several hundred rules in each is a far more palatable thought.
IMPLEMENTATION: This is where things get really exciting. We now have the capability to create true decision services (relying heavily on the power of service –oriented architectures) that are independent units of business functionality. These are transparent and can be made accessible across multiple processes in the Enterprise.
MANAGEMENT: We know a primary benefit of business rules is agility. Business units can define, refine and truly own the business logic they know. By segregating rules into decision, we can uniquely create management requirements for these sets of rules. This categorization by decision allows us to evaluate the velocity of change, scope of change, and the roles necessary to handle change. With that in mind, specific management capabilities can be established on a decision by decision basis.
While the merging of process and rules is clearly a fundamental concept, viewing their integration from the perspective of a decision can be an extremely powerful approach to truly harness and apply the power of the business.
