The amount of information available has grown exponentially in recent years, resulting in oversized data sets – commonly referred to as Big Data. This data explosion has made traditional management tools and techniques for managing information almost incapable of producing meaningful results quickly. This Big Data problem is both significant and widespread.  Analysis that used to take some organizations minutes now takes hours to complete, assuming that one can receive results at all. With the sheer volume of data (much if which is unstructured) that some organizations have amassed, has created a situation where new and more effective methods to identify relevant data and to extract actionable insights out of the most pertinent information possible is required.

However, simply having access to more data does not necessarily equate to more insights. In order to provide business value, insights must be extracted, distilled and analyzed from the raw or aggregated data source(s). This is where Big Data meets Advanced/Predictive Analytics. Predictive analytics, at its core, is really just the understanding the key relationships between business KPIs and past events that dictate a certain set of variables. Understanding these past events allows an organization to predict, given a selected set of external or internal drivers, a probable outcome.

Predictive analytics is all about identifying patterns in the past data, melding them with current data points that are now readily available using today’s technology, and then taking action to ultimately improve business performance. Organizations can profit from predictive analytics, when they ensure that the analytics is based on enough historical data to determine meaningful patterns and the analyst has the discipline to properly model and analyze data for decision-making support. This approach helps turn poor business decisions, made using hazardous guesswork, into well thought out and successful business decisions that improve performance.

An organization must have sufficient and relevant data over time to establish logical data behaviour patterns. This means data that is in a proper format, granularity and is stored in such a way that it can be mined. Simply having large volumes of data, without effectively capturing relevant history, will not provide enough measurement points to detect a rate or change of rate. This approach helps turn poor business decisions, made using hazardous guesswork, into well thought out and successful business decisions that improve performance. However, CFO’s should not immediately jump onto the “predictive analytics” bandwagon before their organization is ready. In fact, just recently Capgemini talked with a CFO who said it best—”I can’t predict my organizations’ financial future without knowing and truly understanding the past.”