Sep 03 2015
For a long time, data analytics broadly meant descriptive analytic, which provides a report of business events that happened in the past. In sharp contrast, predictive analysis provides a futuristic report of business events likely to happen. Thus, the shift in review timeframe from descriptive to predictive analysis indicates that modern business owners are not just happy designing operational strategies based on past events; they want to forecast the future, which is primarily based on the theory of probability, which is the core concept in forecasting. Thus predictive analysts cannot confirm future events, but they can simply forecast events likely to happen.
Regular predictive analysis can build predictive models at the macro level, but big-data enabled predictive analysis can build models on a micro or on an individual level. iNostix provides an example of a big data enabled, predictive analytics solution for the HR industry.
During the telephone interview between Information Week and Dr. Michael Wu, chief scientist of San Francisco-based Lithium Technologies, Dr. Wu explains how descriptive analytics differs from predictive analytics, and what is the business rationale of shifting towards predictive analytics as this type of analytics is perceived to provide enhanced value to organizations.
As Dr. Wu explains in a nutshell, “Once you have enough data, you start to see patterns. You can build a model of how these data work. Once you build a model, you can predict.”
Dr. Wu also noted that while the main purpose of descriptive analytics is to report, the purpose of predictive analytics is to forecast. In descriptive analytics, most metrics are nothing more than event counters, but in predictive analytics, the metrics derived from statistical calculations, modeling, data mining, or machine learning—all combine to form historical trends and patterns to enable analysts to make predictions about the future.
A Datafloq article seems to suggest that whereas predictive analytics is about the future of business performance, prescriptive analytics taken a step forward to provide to offer prescriptive actions to mitigate the adverse effects of future business events.
According to an EMC blog post, BI falls under the realm of descriptive analytics that is simply useful for gathering factual reports of existing events in business, whether past or current. EMC notes that as predictive analytics has the power to look into the future providing insights into the business, this kind of analytics can be especially useful for developing business strategies. Predictive analytics, can easily deliver models of probable business behaviors or events at the micro levels—at the individual customer, product, or campaign levels.
Thus, one may conclude that predictive analytics enable enterprises to move beyond the answers generated by BI, and explore probable actions or events likely to take place in future. That kind of knowledge then sufficiently prepares the businesses to plan for any negative situation that may affect the business or exploit any favorable situation for lasting business benefit.
If now you take a relook at the logical progression of analysis from descriptive to predictive analytics in businesses, it will be easy to understand why the move from de6csriptive to predictive analytics was natural.
Dig deeper into the world of data