When considering which Decision Support tools are needed to meet organisational objectives, you need to consider far more than just the tool itself in order to ensure successful outcomes.
The ‘Decision Support Triangle’ is one way to visualise the key factors that are essential to ensure that the outcomes of Decision Support Tools meet organisational objectives. The three key factors are:
- Data of known quality – First and foremost, you need to ensure that you have data of known quality. By that, we don’t mean that all data has to be ‘perfect’, but instead that the completeness, accuracy and validity of the data is understood. If your data is less than perfect, then it is important to know the nature of these imperfections in order to ensure that both the analysis process used, and the interpretation of the outputs, take sufficient account of the nature of the input data
- Analytic Tools – There are many decision support tools available ranging from locally developed spreadsheet analytic tools through to Monte Carlo simulations and rules-based, artificial intelligence engines. These can be hugely sophisticated analysis tools that require much experience to be able to utilise effectively
- Subject Matter Expertise – In order to be able to develop good outputs, it is essential that a suitable level of Subject Matter Expertise is available to the analyst. This is required in two main areas:
- To ensure that input data is treated correctly and that the assumptions made in the analysis are appropriate based on the quality and completeness of input data
- To undertake both quick ‘sanity check’ tests of the outputs of analysis in order to quickly identify incorrect analysis, and then to evaluate the business impact of the analysis outputs. Does it “tell the right story?” Are the outputs suitable to support follow on business activities? It is likely that the knowledge required to provide this subject matter expertise may exist in more than one individual and may require a small team
These three main factors will help ensure that decision support activities are effective. If you have known weakness in one of these three factors, then the remaining two factors will need to be stronger than would normally be the case. Have I missed any main factors?