I don’t use the question loosely or as a quiz. I really just want to understand what the other person means by “Analytics,” because I want to know where they are coming from. Here’s what I’ve discovered.
Data Aggregation and Descriptive Reporting: Some also refer to this as Data Engineering which is a good term since it requires some of the same qualities as a Data Scientist (a term worthy of a complete other article). This basically means looking at transforming data, creating reports and describing “what” has happened. Our profile reports and most of our client reporting falls into this category. Data engineering is also much more than this, but it’s one example of where that term fits.
Insight Generation: This is where statistics start to come into play and where we provide a lot more value to our clients. In many ways this is the “what is significant with the data” and is often followed with recommendations and/or changes to a strategy. In this category we are defining to answer the “why” something has happened. Still in a descriptive category though teetering on the prescriptive category of analytics.
Predictive Analytics: This is what I mean when I say analytics. I’m not saying I’m right. It’s just the legacy of why Allant’s Analytic Solution team was put in place. A distinct capability of predictive analytics / modeling is to use past behaviors to predict future behaviors or to identify and classify audience segments. Very rarely when I ask people what they mean by “Analytics” is this their answer. But again, that’s OK! At Allant we use several different predictive techniques and advanced analytic tools.
The diagram below defines the “Analytic Spectrum” using several of the above themes. This is one of my favorite diagrams mostly because it has the “looking back and forward” approach. It outlines phases of turning data into information, identifying insights, making predictions and then optimizing those predictions. This aligns to our approach once we layer in a firm understanding of our client’s business challenge and build out a roadmap for continued learnings. That’s what I mean by Analytics!