Predictive analytics have changed the game for such heavyweights as Match.com, Netflix, Best Buy, and the NBA.
And credit unions have begun to harness the power of predictions, too, according to Simon Gao, vice president of consumer analytics for CUNA Mutual Group.
Predictive analysis allows credit unions to better targeting their mailings, route online advertising through favorable IP addresses, and even determine proper levels of call center staffing for the inevitable peaks and valleys.
“A little predicting goes a long way,” Gao said Wednesday during a breakout session at the National Association of Credit Union Service Organizations’ annual conference in Lake Buena Vista, Fla.
As opposed to forecasting, which measures data in the aggregate to determine overarching trends, predictive analysis sifts through the mountains of consumer behavior data that has become available in recent years to determine patterns for individuals’ future actions.
It’s a tool that increases an organization's efficiency—often considerably—and therefore boosts odds of success in a competitive business climate.
“It’s not a crystal ball—it’s not going to change things overnight,” Gao said. “But’s an additional tool that businesses can use to impact their decisions.”
Predictive analysis is highly actionable, he added. “It’s not like forecasting; the predictions are individual. So you can have an action plan to impact the outcome.”
Consider Netflix, which initially provided DVDs only through the mail—a business model that might not have survived long-term against the established video giant Blockbuster.
Predictive analytics enabled the company to tailor suggestions for movies to its customers, buying enough time to refine and launch a streaming delivery channel that eventually put its rival out of business.
Match.com says it has a 95% success rate at determining whether a relationship will stick based on just 10 data points. Best Buy determined that 7% of its customers accounted for 43% of purchases. Not only can the company cater offers to that loyal subset, it can predict which potential customers’ behavior mimic that of the 7%.
And some NBA teams, following the lead of the “Moneyball” numbers-crunching revolution in baseball, have found success using predictive analytics to set the best personnel matchups against a given opponent and finesse game plans.
CUNA Mutual Group has put several predictive models into practice, Gao said, and the possibilities run rampant for credit unions.