TUCON: Design for People, Build for Constant Change

Connie Moore kicked off the Process Improvement track with the Forrester message "design for people, build for change" and dynamic business applications to a packed room. Check out my coverage of her keynote from the Forrester technology leadership conference last year for some background to this theme.

She discussed how methods of working are changing to put the worker at the center, with access to their information, processes, functions and other components as required: the modern information worker decides what he needs to complete any given task. In order to accommodate this, workers need dynamic applications that provide a highly-contextual dashboard/portal interface that might include client information, a calendar of events related to that client’s data, what-if tools for financial analysis, tools such as online enrolment for selling additional products to the client, and other information that’s related to what’s happening right now, not static information.

She sees BPM as going mainstream, and dragged out the hockey-stick growth predictions that all the big analysts love; I’m still seeing a lot of niche and departmental applications of BPM and think that these growth projections may only be met if the analysts continue to change the boundaries of what is considered to be BPM.

She covered several of the reasons for deploying BPM, and walked through some best practices for getting started:

  • Start with a major process that is causing pain: there will be less resistance to change, and easier support and funding. Typically, these are customer-facing, high-volume processes with lots of steps and handoffs. I’m also a big fan of this approach, since no one ever justified enterprise-wide deployment of BPM by doing a proof of concept with managing expense reports.
  • Look for quick hits, using an incremental approach and targeting 3-month release phases. I’m also completely behind this idea, and always recommend getting something simpler into production sooner, then adding on functionality and fine-tuning processes incrementally. I’ve found that BPM implementations lend themselves particularly well to Agile methodologies.
  • Design for real-world processes by doing effective process discovery: avoid interviewing the managers and reading the out-of-date procedures documentation in favor of talking with the people who really know how the process currently works and where the pain points are that need fixing. You don’t want to get too granular here, but use some process modeling tools to sketch things out and identify subprocesses and services. I’m going to expanding on this topic tomorrow in my breakout session, Using BPM to Prioritize Service Creation.
  • Link BPM and SOA. 71% of large companies surveyed by Forrester said that SOA was very important to their BPM efforts: the availability of services is what makes it possible to create and modify processes quickly and easily.
  • Keep the financials in mind. Link projects to the line of business rather than infrastructure, and don’t burden the first project with the infrastructure cost. Measure the results and ROI to use for future project justifications. For ROI calculations, she listed conservative estimates of saving 30-50% of clerical workers’ time, and 20-35% for knowledge workers, with transaction-focused processes seeing even greater benefit.
  • Develop a competency center from the start, including a cross-functional and collocated team of developers and business analysts, strong involvement from the vendor, and judicious use of systems integrations for specific targeted parts of the project. Forrester has seen a strong correlation between the existence of a competency center and measurable benefits in BPM projects.

She recently interviewed a financial services client of TIBCO’s, and they shared a few of their lessons learned:

  • Reengineer the process first, then pick the tool
  • Set the tools aside and focus on the process
  • Be prepared for staffing challenges
  • A competency center is critical

This was really a whirlwind tour of Forrester’s view of BPM, much too much information for a 50-minute presentation but lots of good stuff in here.