What Makes a Great Analysis?
Computing statistics, writing a report, and applying a modeling algorithm are each only one step of many required for generating a great analysis. There is no “easy” button that lets you take one simple step and get a solid result. Not understanding and focusing on what is required to do an analysis right can cause a lot of pain, lead to wrong decisions, and generate enormous levels of extra work.
This chapter will explore several themes. We’ll start by clarifying a few definitions, and then we’ll discuss a variety of themes that relate to creating a great analysis. Each theme will contain a lesson in the nuances that separate reporting or statistic generation from analysis, as well as meaningful analysis from useless analysis.
The principles discussed apply broadly and are not specific to big data. However, with big data adding even more complexity to the mix than organizations are used to dealing with, it’s more crucial than ever to keep the principles in mind. Your organization won’t be able to tame the big data tidal wave by reports alone. Nor will you be able to tame it through substandard analytics.
Too many organizations mistakenly equate reporting with analysis. That may seem harsh at first glance, so let’s clarify what is meant by the statement. Reports are important and can be valuable. Reports used correctly will add value. But reports have their limits, and it is important to understand what they are. ...