The development of analytics-based models for business productivity improvement or optimization has always relied on the analysts understanding of both the data and the tools used for model development. Analytics are the product of a series of iterations in developing theories based on the available variables, seeking correlation or even causality, and then refining the variable set. This iterative nature of the analytic process is too often impacted by the production cycle time for model development, and when the lions share of time is spent wrestling with the tool, it decreases the time spent in analyzing and building effective models.
