Performance Management Coach and Decision-Making Facilitator

In a normal data set information is twice or even more times present. Principal Component Analysis (PCA) is a method that can be used to reduce the number of KPIs of a data source and reduce it to what is really needed. It is an exploratory multivariate data-reduction technique that can be used to simplify complex data sets. The technique identifies correlating KPIs and eliminates overlapping information. It will pinpoint which KPIs are most typical and also the number of KPIs that explain the bulk of the variance. Hence it shows the things that are really necessary and relevant i.e. the needed and sufficient KPIs for proper decision-making.

The principle of needed and sufficient information is valid. Not too much, not too little, just enough. Besides an optimal number of measures the degree of correlation that is found amongst the metrics can be plotted in relation to one another. In certain cases this information might be useful in the search for possible drivers of performance.

The needed and sufficient number of metrics …