Predictive analytics is an interesting buzzword (phrase) created by adding two buzzwords – Predictive and Analytics. The cover does justice to the book here. Predictive Analytics incorporates the use of analytical skills like statistics and machine learning to analyze the available information and existing knowledge, to deduce predictions about future events and forecast outcomes. The goal is to help the business understand how their present and different courses of action would impact their future.

To make more accurate predictions, businesses collect both structured and unstructured data from different sources. After the data is collected and noise is eliminated, using statistical tools and modelling, from the historical data specific patterns are identified which lay the foundation. Tools like AI and ML help in leveraging these patterns in conjunction with current circumstances to forecast the future.

The success of Predictive Analytics lies in how accurate the predictions are. For example, based on historical data and recent sales, a t-shirt manufacturer decides to manufacture a volume of 10 thousand t-shirts of a certain type for the holiday season. This means investment in raw material procurement, manufacturing and warehousing expenses and finally logistics expenses. If the predictions are inaccurate, not only would a part of the investment be wasted, but there will be a loss of opportunity in terms of shifting focus to some other segment.

Predictive Analytics is used in every domain and industry. Primary applications are in Retail and Supply Chains where patterns are repetitive and data collection sources are wide across the board.

~S