Here in Arizona, you can still find folks prospecting for gold. Some chose to pan for gold in streams and other try to study the geological patterns and look for quartz deposits (quartz and gold tend to be found near each other). Both have been successful in locating gold. Nobody has unraveled the history behind the “Lost Dutchman’s Mine” enough to find it, however. Fortunately it is much easier to find process benefits that it is to find gold, though the value is great in either case.
![MMj02852560000[1]](http://blogs.gartner.com/jim_sinur/files/2009/05/windowslivewriterchooseyourprocessdiscoverymethodmodelorm-eb44mmj028525600001-2.gif)
Process Modeling:
The nice thing about process modeling is that it is fairly easy to start with a focused process discovery exercise that brings key folks together with a facilitator to look for nuggets of benefit. All you need is a blank wall and sticky notes and/or a blank writing board. While a modeling tool can help, it is not necessary to find low hanging fruit in the exiting processes A simple model can then be simulated and/or iterated (See http://blogs.gartner.com/jim_sinur/2009/03/24/simulate-versus-iterate-mutually-exclusive/). The issue with modeling is to know when to quit and how to get the right folks there with a methodology driven and experienced facilitator.
Process Measurement:
While process measurement usually requires some kind of automated business process discovery tool (ABPD), the accuracy of the “as is” model is quite helpful in locating opportunities. The payback is quick and ABPD can use readily available logs, application systems and activity on the glass of a connected device like a PC. The ABPD technologies are good a visualizing the behavior of many of the process resources (See http://blogs.gartner.com/jim_sinur/2009/03/12/automated-business-process-discovery-helps-visually-optimize-processes/)
This is not a zero-sum situation. One can use both, but it is rare to start out with both approaches. One eventually leverages both. Keep in mind that process discovery is only one of many places to start a BPM effort (see http://blogs.gartner.com/jim_sinur/2009/04/14/the-top-five-bpm-starting-points-today/).
2 responses so far ↓
1 Robert Shapiro // May 27, 2009 at 2:30 pm
In respect to process modeling I suggest that anyone who plans to do it should check out the new Book by Bruce Silver: BPMN Method and Style, http://www.bpmnstyle.com. Bruce has a very different take on explaining BPMN based on teaching thousands and thinking hard about how to be clear and concise in modeling business processes.
For analytics and process discovery take a look at the work being done at Einhoven University on the ProM framework, an open source project with a lot of tools in support of process discovery and analysis.
2 Jacob Ukelson // Jun 1, 2009 at 4:38 am
Jim,
Another way to start understanding existing processes is to look at existing unstructured processes in the organization. The problem is that most of these are done via email and documents in an ad-hoc fashion – making automated discovery very difficult, if not impossible.
One way to move forward is to put a simple amount of structure around those existing unstructured, ad-hoc,human processes that woudl allow for automated discovery -but still let people work as before.
If anyone is interested, I have written some posts about this in the past in our blog:
http://blog.actionbase.com/in-situ-process-discovery
and
http://blog.actionbase.com/actionbases-third-way-using-an-hpm-to-easily-turn-an-organizations-practical-intelligence-to-explicit-knowledge
Leave a Comment