I am attending Gartner Symposium and ITxpo this week and blogging as I go. Last session before I have to leave for the airport.
BPM requires a methodology or approach as well as technology but this about technology. In business today Six Sigma and Lean Six Sigma both increasingly used and both focus on continuous improvement and BPM can really help develop this. Jim outlined the development of BPM from just modeling procsses, to modeling and executing, to integrating people and systems and monitoring all of this. The market is now seeing what Gartner call a BPM Suite with new features include optimization, in-line simulation, business rules, business rules management and scenario management for rules.
Jim talked about the growing intersection of rules and analytics, what I call Enterprise Decision Management or EDM, and I could not agree more. He also talked about goal-directed BPM. I already see some work around goal-directed decision-making (to come up with optimal decision rules) and this could certainly drive goal-directed processes. Business people meanwhile will become more and more involved in the definition and simulation of processes (and rules) and the advent of widespread SOA will reinforce BPM very effectively - I wrote about the intersection of BPM, SOA and CEP before. In this vision BPM technology composites and orchestrates services to create business value and cut across silos. Jim talked about this view of composite applications and the widespread use of rules to add value to standard components. This is where all this technology goes.
Well that's, I am out of here. Back to regular blogging next week.
BPM Suites make lots of things explicit:
- Processes becoming explicit and being managed more and more by business professionals. Jim talked extensively about using a business rules engine to deliver decision-agility to the process without having to change the process. I wrote some stuff on Gartner's approach to agility and how to achieve it that includes consideration of both rules and process management for agility.
- Document and imaging routing becoming explicit so as to make documents available in a process context.
- Modeling makes the processes and their dependencies explicit so they can be simulated and managed. People really want a common modeling approach and a way to share models but nothing yet arrived.
- Integration brokers make system to system interactions explicit and makes data and transactions visible. Most of these lack rules as part of this - like webMethods for instance, who OEM Blaze Advisor.
- Portals make user navigation explicit and manageable and can guide best practices and bring multiple work threads together.
- Rules to manage decision and diagnostic logic - need rules in services, integration, decisions, more places. These rules are easier to share and change and so on. I wrote about different ways to build decision services and to use rules before. Jim also feels that rules are the right way to help someone take action when dashboards tell you that somethings up. I wrote about this in the agility piece above and in an article here on putting performance management to work.
In terms of why use technology of this kind he made some points:
- System change requests spiking but only about 35% of rules are changeable enough and about 20% of them will be managed by IT leaving 15% to be managed by the business. Personally my experience is that everyone ends up with more changeable rules that they expect.
- Exception handling must be improved and thus people must be managed as part of this
- CRM interactions, both self-serve and customer-service processes are more and more key
- Compliance and oversight imposing constraints that must be managed without a loss of agility
- Outsourcing will be easier as you can control the critical decision nodes within a process
- Business users must control more
In the future you see
- BAM or Business Activity Monitoring (I blogged about EDM and BAM)
- Smarter user interface components (SmartForms perhaps)
- Pre-built components and templates (like Fair Isaac's Decision Accelerator for origination perhaps)
- Predictive modeling (presumably injected into processes by analytically enhancing rules-based decisions)