In a recent Datamation article, "Business Intelligence Software and Predictive Analytics," Jeff Vance asks, "What exactly is predictive analytics?" His answer, in part:

"Take traditional business intelligence, combine it with data mining and add on statistical analysis and you have predictive analytics. Math geeks will squabble over the nuances, say, whether a specific model is a predictive, descriptive or decision-making one, but for most organizations this boils down to using historical data along with probabilities to better assess the future." 

Of course, I've already shared (and maybe over shared) Tonya Balan's definition of predictive analytics in the sascom article, "Understanding Analytics," so it's nice to see Jeff also quoting her simple example:

"Tonya Balan, manager of the analytics product management team at SAS Institute, offers an example of how predictive modeling is different from simple forecasting. Forecasting will tell you that you’ll sell more ice cream cones in July than other months of the year." 

I've also had some interesting conversations lately with experts here at SAS like Anne Milley, Keith Collins and Mark Torr about the difference between analytics and predictive analytics - and why BOTH are important for businesses.

We created this fun, little article a few years ago that described the 8 Levels of Analytics, which was popular and shared around quite a bit online. The evolution images also get used here at SAS a lot in presentations. In general, the last three images depict predictive analytics.

When Keith uses these images, though, he turns them around and starts with the optimization guy first, because you don't actually have to do them in order. You can - and should - start with whichever type of analytics is best to solve your specific business problem.

Likewise, when Mark and Anne talk about analytics, they talk about business problems, and how you might need forecasting for your supply chain and query drilldown for merchandise planning. Anne likes to describe how you can start with one type of analytics and build on the next, most logical, type of analytics for your business (again, not necessarily in the order shown here). You may, in fact, need OLAP and forecasting for your merchandise planning project.

The three people I've mentioned here are just a few of the many employees at SAS who know business analytics technology inside and out - and also have a knack for explaining it to less technically savvy people like me.

If you like to talk about analytics, you should join us in Seattle next week for SAS Global Forum. I'll be there. Keith and Anne will be too. We'll be talking a lot about analytics, and I'll be live blogging as many presentations as I can. We'll even be talking about another topic Jeff mentions in his Datamation article: Using Social Media to Glean Predictive Analytics.