Better Processes Start With Better Measurement

To initiate corrective actions dynamically, the BPM software must be able to correlate events in disparate segments of an overall process or in multiple processes. It also must be able to gather and manipulate external data and combine it with process data for use in dashboard reports and business rules.

Calculating the total cost of each instance of a function such as order fulfillment, for example, means adding up all the costs of activities in multiple workflows. Performance management should be able to collect data across all activities, aggregate costs for each order and present the results in dashboard reports. Ideally, the tools to define the data, report layouts and triggered actions shouldn't require programming or even database work.

BPM software varies in performance-management capabilities and approach. Some products rely on an embedded business-rules engine working in parallel with the process-automation engine. As the process engine routes instances through a workflow, the rules engine listens for events, generated by the process or by external systems. The rules engine can update performance metrics based on those events, issue a new event tied to those metrics or invoke actions directly on the process engine.

Other systems build steps into the process model itself, logging data at specific points in the flow to a tracking database designed for real-time reporting and online analytical processing. Again, when metrics for the process as a whole track out of a target range, the software issues real-time alerts or takes remedial action automatically.

Whether event-based or built into the flow model, active performance management in BPM is giving process-centric organizations a critical new competitive advantage.

Bruce Silver is president of Bruce Silver Associates (www.brsilver.com). Write to him at [email protected].