Top Down System and Data Discovery

Operations management focuses on the management of the resources required to produce a service or a product. These resources typically consist of people, materials, technology, and information (Rowbotham, Galloway, & Azhashemi, 2007, p. 8), primarily through the employment of dynamic systems. These systems can be viewed as elements working together in pursuit of a common goal (Haksever & Render, 2017). More specifically, these systems can be viewed as a productive system due to the transformation of system inputs into value added outputs. These systems are controlled through the insertion of feedback and control mechanisms that allow for the diagnosis and correction of the inputs, outputs, or the transformative process.

While simplistic on its surface, an organization typically consists of many systems. These systems may receive inputs from external sources, and provide outputs to another external source, receive inputs from external sources, and provide outputs to an internal source, receive inputs from internal sources, and provide outputs to another internal source, receive inputs from internal sources, and provide outputs to an external source, systems may provide inputs to other systems, to produce secondary and tertiary outputs. Each system may also be further decomposed to provide details to the transformative processes at their very heart.

At the highest level, an organization (or subset of an organization) tasked to provide transactional service to a customer could be represented by the above image. However, a closer read of the contract shows three types of transactional processing (variety), as well providing for contact center services. Decomposing the transactions, we find roughly 52,000 annual transactions (volume), made up of about 300 distinct types (variation), from three main input sources – each made up of multiple individual sources. Further complicating our original view of the system are over 27,000 contacts in the contact center, each of which may create new transactions or modify existing transactions already underway (customer contact). These elements (volume, variation, variety, and customer contact) define the environment in which the system operates. Volumes speaks to the number of times a system transform inputs to outputs. Variation describes patterns in the volume. Variety is the number of different services provided. Customer contact describes the amount of time spent interacting with customers (Rowbotham, Galloway, & Azhashemi, 2007). Additionally, and more frustrating to those tasked with the management of these systems, when poorly designed or inefficient systems are accepted as business-as-usual employees begin to create their own unofficial (and often undocumented, and uncontrolled) workarounds to circumvent the inefficiencies (Marquiss, 2009).

With all these processes running, some serially, some sequentially it should become obvious that there must be a consistent and routine analytical structure embedded in these systems to provide the mechanisms for management to ensure both the efficient and effective operation of the systems. When embedded into the organizations systems, the analytical measures are no longer ancillary or superfluous, they instead become consistent, routine, and natural parts of the day-to-day activities of the organization (Davenport, Harris, & Morison, 2010, p. 117).

To become ingrained, and to remain useful, the measures must be focused on how the organization defines value and within the given context of the system itself (International Institute of Business Analysis, 2015). Often these measures are influenced, and possibly imposed by contractual or regulatory specifications. Contractually these are referred to as service-level agreements or shortened to SLA. The SLA is a commitment between a service provider and a client. The SLAs will define the aspects of the services provided (e.g., quality, availability, responsibilities) (Kearney & Torelli, 2011).

The SLA provide a starting point for the development of system performance measures but provides limited insight. Like the processes they represent, these measures need to be decomposed to provide more visibility into the process to develop a better sense of the details behind the measure. A simple percentage (see Figure 2) contrasted with the stated SLA is often used to report on SLAs.

While a continuous reporting on the percentage can indicate progress towards an SLA, it proves inadequate to gain insight into the events that led this measure to reach 57%, it also does not provide any insight into the level of effort, or the potential issues remaining. As illustrated by the red percentage and the addition of the exclamation point, additional insight is added immediately to draw attention to SLAs at risk. 

To provide the simplest of answers (57% of how many? How many are left?) additional detail must be provided. Figure 3, below, shows the same SLA with additional detail provided. With the additional data it is now possible to see that 57% of 333 possibilities have been closed (using the common standard of green is good, red is bad), it allows the reader to determine the number of closed (189) and open (144), the number remaining that must be closed to meet a specific threshold (115), the number remaining that must be closed to hit the SLA (128) and again that there are 144 remaining.

Figure 3 Expanded SLA

Reviewing this measure, it is easy to guess the follow up question. Given that we have 144 more how close are we in getting those done. To answer this, it is required to continue to decompose the data. As with the processes above, the further you investigate the data the more you can distinguish. Figure 4, below, shows the same data point, but at a lower level. In this visualization it is possible to see the work in progress and the number at each of the 6 stages in the lower level process.

Figure 4 Decomposed

As with the systems above, this decomposition can continue, and more and more measures can be developed to provide other insight. The trick is in decomposing your systems to the point that you have adequate measures and developing those measures to answer the specific need at that level.

References

Davenport, T., Harris, J. G., & Morison, R. (2010). Analytics at Work. Boston, MA: Harvard Business School.

Haksever, C., & Render, B. (2017). Service and Operations Management . Hackensack, NJ: World Scientific.

International Institute of Business Analysis. (2015). A Guide To The Business Analysis Body Of Knowldge. Toronto: International Institute of Business Analysis.

Kearney, K. T., & Torelli, F. (2011). The SLA Model. In P. Wieder, J. M. Butler, W. Theilman, & R. Yahyapour, Service Level Agreements for Cloud Computing (pp. 43-68). Berlin: Springer Science+Business Media, LLC.

Marquiss, R. W. (2009). How To Gain Better Operational Leverage Through Enhanced Process Maturity. Cognizanti Journal, 42-48.

Rowbotham, F., Galloway, L., & Azhashemi, M. (2007). Operations Management In Context. Oxford: Butterworth-Heinemann.