Analytics Mistakes (and how to avoid them)

Robert Mitchell discusses 12 mistakes associated when undertaking an analytics project. Any student of Principles of Marketing should recognize hist list because it is very similar to the list for analytics successes from the last chapter in the Principles of Marketing Engineering. Of his list, I cringed when I read number 6:

Don't just proceed, but rush the process because you know your data is perfect.

I recall two more recent incidences where I was given a dataset. In the first instance, I did no validity checks or run descriptive statistics. I lost roughly four months in attempt to understand why the data was producing results inconsistent with theory. If I had performed checks and run descriptive statistics, then I would saved myself three months.

In the second instance, I did perform the necessary checks and statistics. I found several issues with the data, which altered my initial thoughts of the analysis. I developed a different model and, as a consequence, the analysis was completed in about a month.