Data Mining explores, predictive analytics answers "What next?"
Predictive analytics and data mining both apply sophisticated mathematics to data in order to solve business problems. But when people talk about data mining, they are usually referring to an analytic toolset that automatically searches for useful patterns in large data sets. By contrast, predictive analytics is an analyst-guided (not automatic) discipline that uses data patterns to make forward-looking predictions, or to make complex statements about customers by evaluating multiple data patterns. If data mining searches for clues, predictive analytics delivers answers that guide you to a "what next" action. Data mining is often one stage in developing a predictive model. Automated data mining techniques can isolate the most valuable data variables within a vast field of possibilities. The analyst uses those variables, and the patterns those represent, to build a mathematical model that "formalizes" these relationships and predicts future behavior consistently.
BI delivers insight, predictive analytics delivers action
Traditional business intelligence (BI) tools extract relevant data in a structured way, aggregate it and present it in formats such as dashboards and reports. Like data mining, BI tools are more exploratory than action-oriented, but the exploration is more likely driven by a business user than an analyst. BI helps businesses understand business performance and trends. Whereas BI focuses on past performance, predictive analytics forecasts behavior and results in order to guide specific decisions. If BI tells you what’s happened, predictive analytics tells you what to do. Both are important to making better business decisions.
Predictive analytics also focuses on distilling insight from data, but its main purpose is to explicitly direct individual decisions. Many BI suites now include some analytics, ranging from report-driven analytics that synthesize past performance data to predictive analytics used in forecasting. However, BI analytics almost always aggregates past customer data in a collective sense—for example, how have my customers behaved so I can forecast product sales by quarter?