The chart above offers a visual of 4 types of analytics from least complex rising to most complex (Descriptive analytics to Diagonstic analytics to Predictive Analytics to Prescriptive Analytics). Descriptive analytics answers the question of what happened. Raw data from large data sets is used to look at the past. Diagnostic analytics uses that historical data and measures it against other data to answer the question of why something happened. It helps you drill down or identify patterns. Often you see data visualization used here. Predictive analytics tells what is likely to happen. It uses the findings of descriptive and diagnostic analytics to detect tendencies, clusters and exceptions, and to predict future trends, which makes it a valuable tool for forecasting. But note that the accuracy of forecasting is just an estimate! Accurate forecasting depends on large amounts of quality and consistent data so be careful. Prescriptive analytics is used to advise you on what action to take. Prescriptive analytics requires not only historical data, but also incorporates outside information. This is where things like machine learning, artificial intelligence, and very sophisticated algorithms some into play.
