Dynamic case management (DCM) aims to empower knowledge workers, improve efficiency and deliver better outcomes. Adding analytics to the mix can add even more benefits by providing new insight into case information. But navigating some of the issues involved in combining the two can be challenging, to say the least.
For that reason, experts say it’s critical to embark on such efforts with a clear understanding of what’s involved, as well as a grasp of the limitations. Then apply some commonsense approaches to implementation, with a dose of caution thrown in for good measure.
WHERE THE COMBINATION WORKS
First, make sure the approach is right for your environment and business goals. Keith Harrison-Broninski
, a software architect and consultant, says the combination works well in situations in which individual participants have the expertise and ability to exercise judgment. That includes tasks such as medical diagnoses, mortgage processing and investigating discrepancies on invoices.
However, other tasks involving more complex interactions and decision-making may not be suitable for DCM, with or without the analytics boost, Harrison-Broninski says. Examples of activities that he says probably don’t fit include complex sales, research and development and work related to mergers and acquisitions.
“DCM analytics, such as they are, can help with applying techniques such as Lean Six Sigma to routine human work, but that's it,” he says.
Keith D. Swenson, co-author of "Mastering the Unpredictable:
How Adaptive Case Management Will Revolutionize the Way That Knowledge Workers Get Things Done” (Meghan-Kiffer Press, 2010), has a more sanguine view of the potential for DCM.
In Swenson’s view, the BPM wave, which got started around 2000, “ended up being a programming thing” rather than the revolutionary business force that many had hoped for. “Case management came along to fulfill some of that hope,” he says. “The whole idea is still about supporting human organizations.”
BPM VS. DCM
Swenson finds it useful to contrast the application of analytics in DCM and BPM. In BPM, he says, the central idea is getting to a perfect process. “If you go and talk to everyone in an organization you can define that perfect process and say, this is the best we can make and everyone should do it this way because it is better than simply being random,” he says. “In that BPM scenario, analytics is used to ensure conformance to that process.”
However, with DCM (and with knowledge work in general), you should avoid defining processes too tightly because they often involve unpredictable factors. “If you just try to make up a process [that way], you will probably end up with a process that is dysfunctional,” he says.
He compares the approach to what’s involved in conducting a criminal investigation, when detectives compile information on the crime, following clues as they come up. “There’s no way to say ahead of time ‘This is the way to do it,’” Swenson says. “It’s like Sherlock Holmes: You never know what clue you’ll get.” A successful investigation requires enough flexibility that the knowledge workers—in this case, the detectives—can do what they feel is necessary to achieve their goals—in this case, solving a crime.
After they’ve solved, say, 100 crimes, they can examine what happened and learn more about their procedures through a form of analytics known as process mining. “It takes in a bunch of log data and, from an aggregate across cases, it will produce a process diagram,” Swenson explains. “It is pretty amazing.”
However, he cautions, process mining won’t spit out a perfect five-step, one-size-fits-all diagram for solving crimes. “The reality is that different detectives do the work in different ways,” he says. Still, process mining can provide plenty of insights about the most effective investigation practices as well as revealing patterns of criminal activity.
BEGIN WITH BASICS
It can also be helpful to map out basic tasks that are essential to the case process, says Scott Cleveland, who writes the "BPM from a Business Point of View"
blog for ebizQ. Those steps can be mapped out through a BPM system with various additional exceptions provided at each step so that a caseworker can add activities and people to the process, says Cleveland, director of sales and marketing at Impac Systems Engineering, which provides engineering design and software integration services. Harnessing the BPM software for a DCM task can provide a trail of how the case was handled and moved through your organization, he says.
No matter what method you choose for implementing case management and applying analytics, DCM initiatives should always be placed in close context to a business problem you are trying to solve, says John Lucker
, a principal and leader of the Advanced Analytics & Modeling National Practice at Deloitte Consulting LLP.
“Exactly what the best practices are and how you apply analytics with dynamic or adaptive case management will vary depending on the business application,” he says. For instance, business problems could be as diverse as medical case management, insurance claims or fraud detection—and could also require real-time or near-real-time response, he notes: “Those factors will vary the type of DCM response you develop.”
Are you using analytics in dynamic case management? If so, how did you decide that the combination was the right one for your business environment and goals? ebizQ's staff would like to hear about what factors you considered. Please e-mail Site Editor Anne Stuart at firstname.lastname@example.org.