Article Preview
Top1. Paper Overview
In the COBIT 5 Framework for IT Governance, the Goals Cascade is positioned as a key construct, aimed at putting strategy in practice by prioritising IT enablers, including IT Processes.
This research is an answer to De Haes et al. (2013) in which a call is made for more research on COBIT 5 as an artefact, and hence we will research the Goals Cascade, more precisely whether it is fit for its intended purpose.
In our literature review we found references to the use of the goals cascade and some case studies where the goals cascade was applied. However, no authors challenge the cascade results nor did they customise it. Given this research gap and the unchallenged reliance on the published goals cascade, we believe that further research is highly timely and relevant.
We consider the Goals Cascade as an information item, with a declared goal of assisting in prioritising IT processes based on enterprise goal priorities. Researching an information item for its intended purpose equals researching its quality. Via a short literature review on assessing quality of information, we concluded that the COBIT 5 Information Reference Model provides an adequate information assessment model. We used this model to split up our main research question into several sub-questions.
First we looked at the intrinsic accuracy of the published goals cascade, by comparing the results of its application with the application of the original research data. We find that intrinsic accuracy is not very high.
Next we discuss the interdependencies between different goals in the same set of goals. We find that enterprise goals in the Goals Cascade are inter-independent. This has a limiting influence on the scoring of those goals in a prioritisation exercise.
As final step of our review, we have run a number of simulations with the published goals cascade to assess its sensitivity for input variations. We find that both in process weight as well as in relative process ranking the current goals cascade is virtually insensitive to input variations. In conclusion, for the reasons aforementioned, we state that the currently published Goals Cascade isn’t a very good process prioritisation instrument.
In order to resolve these issues, we define a number of basic requirements for a solution, and we propose the introduction of a new construct in the Goals Cascade, i.e. ‘enterprise strategies’, and we defined a limited set of four generic enterprise strategies.
We have performed a proof of concept study for this new construct, including an initial assessment of the viability of the new construct, the initial population of a new mapping table between enterprise strategies and IT processes, and for a high-level assessment of contingencies. The result of our proof-of-concept study is promising, and shows a substantially higher sensitivity for input variations, making it a potentially superior prioritisation instrument compared to the current Goals Cascade.