Blog

For FedEx Day 2014, Cesar, Jack and I worked on a data mining project that investigated sales estimates on various projects at DMC. The result of the day’s work was a report with several insightful metrics, charts and distributions that illustrated DMC’s historical performance in being able to predict the amount of billed time for a project at the quote stage. This report is intended to be used to gauge and adjust future sales estimates/quotes. 

The main lesson we learned is that all the analysis in the world is not worth as much as good record-keeping. It also became apparent that sometimes the metrics you use are dictated by the quality of the data you are analyzing, and the “best” metric to measure something is more of a normative question than an analytical one. Finally, we discovered that accounting has a lot more gray areas than engineering, and I’m glad I’m not an accountant.