By James Taylor, Founder and CEO, Decision Management Solutions
One of the classic tools for decision automation is a business rules management system, or BRMS. While these systems are powerful and effective when used correctly, many companies fail to apply best practices when using them. Here’s one of our favorite horror stories about a well-intentioned BRMS implementation gone awry.
Some years ago, we got a call from a company asking for our help with their business rules project. They said that they had hit a milestone in their rules analysis—it was 80% complete, or so they thought. They wanted to have someone come in from outside to review their rules and see how it was going, so they turned to us for help. We asked how they were gathering and organizing their business rules requirements. They replied that they had a large spreadsheet—organized by types of rules—and that they were recording their rules as they harvested them.
We responded by telling them that they were not, in fact, 80% complete with their rule requirements. Without even looking at their rules we knew this. How? Because this kind of rule harvesting results in a common BRMS horror story: the big, scary big bucket o’ rules. The company had no way to tell if they were really 80% complete. What they meant was that they had used 80% of the business rules requirements budget and they hoped they were 80% complete with the work. But using up a certain portion of the budget doesn’t mean you’ve reached your objective.
How did we save them from the horror of the big bucket o’ rules? Well, we built a decision model of the decision they were trying to make using their business rules. A top-down decomposition of the decision using modern decision modeling approaches and the Decision Model and Notation (DMN) standard showed exactly what decision-making needed to be automated. The rules they had harvested were then reorganized and restructured according to the decision model. As expected, we found they had some sub-decisions that were well defined and complete and some that were mostly complete.
But they had also harvested many rules that had nothing to do with the decision, wasting valuable effort because they had no touchstone for their harvesting work. Plus, there were sub-decisions for which they had done no harvesting. Without a model of the decision-making they had not realized they needed to know those rules. And the lack of a business-driven model meant that many of the rules were too technical, and phrased in ways that made little sense to the subject matter experts.
Fortunately, this story has a happy ending. With the decision model in hand, we were able to help them finish their requirements and the decision model-based rules were easier to implement than they had expected. This enabled them to get back on schedule despite their wasted rule-harvesting effort.
What’s the morale of the story? Avoid the big bucket o’ rules horror. Put decisions first.
Want to learn more about decision modeling? Come to our webinar, “Maximizing the Value of Business Rules” on November 11th 2020. You’ll learn a proven framework that will help you gain competitive advantage, be more efficient, and improve ROI.