Por Kay Winkler


The Association of Business Process Management Professional's Business Process Management Common Body of Knowledge (ABPMP CBOK v2.0) lists process performance measurement as a fundamental element of the BPM practice. Depending on different possible viewpoints within and around the process environment, those measurements may form the foundation to justify a BPM engagement in the first place (project evaluation), identify the ROI of such an engagement – post project or to pursue continued improvements to existent processes.



Establishing viable metrics can at times entail severe challenges. Trying to establish a robust framework of measurable variables that encompass the most crucial business process patterns without causing an excessive accumulation of sometimes-redundant data points, typically requires the design team and the process owners alike to critically boil down the most representative dependent and independent key variables to be measured throughout the company's processes. This definition of measurement variables in turn should then occur, ideally, early in the process design phase and take into consideration challenges like confidential information (example: hourly costs of employees for human centric processes) that likely won´t be captured on the process level.

One of the most common and also intuitive metrics that BPM users, analysts and providers refer to is the process cost measurement. In practice however, especially in human centric processes, sometimes due to the confidentiality of resource cost information and sometimes because of its unavailability at design time (pointing to whole different set of additional challenges), process costs (ex-ante and ex-post BPM) often can't be exactly and easily represented as monetary values (leading to sub sequential and colorful ROI guesswork).

An alternate and eventually more generalized approach to that dilemma maybe the declaration of "TIME" as the principal and dependent core variable for all economic process measurements, which of course will suffice only for service processes whose material inputs play a merely secondary role to its results. Different types of "TIME" can be easily and natively captured by most BPM platforms and later be analyzed, compared within the same company (different processes and versions) and also benchmarked among different corporations. Different economic scenarios, tactics and strategies can then provide monetary multipliers to the continuously measured "TIME" variable, delivering a dynamic business context to process owners and analysts, on demand - without having process costs as an embedded BPM metric.

The core challenge hereto would be the definition and the coherent measurement of different types of "TIME" variables. One can differentiate (among other types) between the task lifecycle time in a given process and the worked time of different individuals for a task during its lifetime. Drilling down further, one could argue it would be necessary to differentiate between the overall task lifecycle time and only its workday lifecycle time (net lifecycle) and so forth.

Having established "TIME" as a dependent core variable for process measurements and the possible formulation of behavioral hypotheses, result-influencing independent variables can now be included into the process metric framework.

For retail and corporate banking processes "net task lifecycle time" as a dependent variable and "quantity of process handoffs," "quantity of monthly process transactions" and "quantity of fully automated core and legacy integrations" – all as independent variables – have shown a significant correlation to each other as a result of BPM engagements.