Predictive Analytics

Predictive analytics is a hot topic right now, driven by the success of professional sports teams, gambling casinos, hotel chains, and banks in using historical data to predict patterns and estimations of the future.

Moneyball: The Art of Winning an Unfair Game by Michael Lewis tells the story of how the Oakland A’s, one of baseball’s lowest payroll teams, could consistently win games and secure a position in the postseason playoffs.

While other teams paid top dollar for top stars, Oakland crunched numbers and analyzed statistics to determine which players were undervalued but likely to perform well.

Wikipedia defines Predictive Analytics as an area of statistical analysis that deals with extracting information from data and using it to predict future trends and behavior patterns.

The core of predictive analytics relies on capturing relationships between explanatory variables and the predicted variables from past occurrences, exploited to predict future outcomes.

Predicative analytics requires a data rich history and environment to employ approaches and techniques broadly classified as regression techniques and machine learning techniques (algorithms and technology to allow software to learn).

Predictive analytics has been part of the scientific community for years and often the basis of scientific insight used to develop a hypothesis.

Card sharks even use predictive analytics in a game of Blackjack.

Based on the cards played (history) in a 52 card deck, the gambler extracts that information to predict the odds of winning a hand based on the future and pattern of the remaining cards.

Customer Relationship Management (CRM) is a frequent commercial application of predictive analytics to extrapolate past consumer behavior to predict future outcomes.

Underutilized in most organizations is the goldmine of data and information available in their revenue, cost, and performance management systems and models.

Customer segments and segment profitability, product and service profitability, channel profitability, plant capacity and utilization, demand curves, and activity consumption rates are just a few examples of this data rich environment.

Applying predictive analytic tools and techniques in this data rich environment of historical performance should significantly improve budgets, financial forecasts, and operating plans.

Use predictive analytics to bring more clarity to your crystal ball…

John A. Miller

December 10, 2007

Leave a Reply