Inefficient at the speed of light
Back to basics
One of the most fundamental knowledge constructs consists of S-P-O, the simple notion of subject, process, and object. However, much of our KM efforts tend to focus on either subjective or objective knowledge. The critical connection between the two, process knowledge— knowledge about the interaction between subject and object—is often shortchanged. Once a process has been put into place, across time, it can take on a life of its own, essentially running on autopilot. Examples include separate reports filed for every meeting a sales rep has with a customer or prospect. Or, worse yet, those dreaded mandatory quarterly and annual performance reviews that nobody reads. Or anything for which management is pushing “100% participation.”
Such processes may be clear, simplistic, and routine, but may not be at all worth the time consumed. The missing knowledge is often the answers to questions such as these: “Why are we doing this?” and “Who cares?” As KMers, the essence of our work is making better decisions. And in today’s extremely competitive environment, that should be a core requirement for each and every process. KM can help break a process-centric mindset by creating a more holistic model in which all three elements, S, P, and O, are present and in balance. Call it knowledge-centric process mining.
An abundance of opportunities for KM
The IEEE Task Force on Process Mining has published a detailed manifesto (tf-pm.org/resources/manifesto) with a set of six guiding principles and five maturity levels. It also breaks down process mining into three stages, each with a specific type of output. The process discovery stage extracts a process model. The conformance checking stage monitors the performance of the extracted model and produces a set of diagnostics. The model enhancement stage results in a new or refined model.
While process mining started years ago as a mainly data-driven exercise, its stated goal is to be knowledge-driven. Given KM’s multidisciplinary scope, we can play a major role in achieving that goal. For example, refining a process model to include the human aspects. Or, incorporating agility and resilience, especially in the presence of complex and volatile global supply chains. Any process, no matter how simple, has the potential to reach across an entire business ecosystem, including all stakeholders. This seems like a perfect match for collaborative workflow, AI/ML, knowledge graphs, human sensemaking, and many of the other arrows in our KM quiver. We can even borrow cyclomatic complexity measurement tools from the software development community, which has been modeling, monitoring performance, and refining their processes by reducing complexity since Day One.
Still concerned about a potential jobs crisis resulting from AI? An exciting opportunity exists for new jobs in deep process mining, helping organizations uncover the rich treasure trove of tacit knowledge hiding deep in their legacy processes. Just because we have the capacity to process unlimited numbers of transactions and reports and to collect and analyze zettabytes of data in real time doesn’t make us more efficient or effective. It may very well make things worse by increasing the odds of making poor decisions based on conflicting, contradictory information.
KM has benefited mightily from text and data mining. It’s time to add process mining to our KM toolset. Think of the flood of creativity, innovation, and productivity that would be unleashed if we identified and eliminated most, if not all, unnecessary, non-value-added processes.