Predicting the Future Is a Strategy, Not a Wish

In a year hallmarked by uncertainty, accurately predicting what may lie ahead isn’t just wishful thinking – it’s an important business strategy. This type of intelligence can separate the best-performing credit unions from the pack in the low-rate environment where greater threats of credit losses loom.

Understanding where the trends are pointing doesn’t come from lucky guesses, gut instinct or feelings based on past experience. The most powerful insights and forecasts are derived from dimensional, data-driven analysis.

The power of predictive analytics is in advanced statistical models and machine learning algorithms that deeply analyze member data and see what the human mind can’t. Especially in today’s climate, credit unions’ interest in predictive analytics is growing, particularly in three key areas: Pinpointing risk, understanding changes in behavior patterns and finding the right places to increase income.

Identify the Greatest Areas of Risk

Many credit unions rely on a combination of intuition and single, specific metrics to assess areas of risk. Analytics-driven models take risk analysis to the next level with processes that are much more thorough, objective – and even more exciting – automated.

A predictive analytics model looks at historical data in multiple dimensions, and then studies the relationships between those dimensions. Models that are powered by machine learning continuously “learn” from the data, and update and adjust forecasting scenarios automatically. This means that predictions are not only more precise and detailed, but they continue to become more accurate and enriched over time.

Credit unions will reap significant benefits by using predictive analytics to forecast delinquencies and charge-offs. This gives the credit union a fuller understanding of its membership and allows segmentation based on risk profile.

Recently at CU Rise, we created a delinquency risk model for a large Midwest credit union with more than 250,000 members. The credit union wanted to build out separate collection models for each of its product types and generate member-level risk prediction. The goal wasn’t just to produce a list of members that were likely to pay late, but to identify exactly who had the greatest propensity to fall more deeply into the delinquency cycle in the next six months. This model will greatly optimize collection efforts, dramatically improving both efficiency and effectiveness.

Understand Changes in Spending Patterns

The pandemic has introduced new unpredictability into spending patterns this year. To be clear, even the best credit union predictive analytics model couldn’t anticipate a global pandemic (not yet, anyway …), but it can continuously assimilate new information to understand shifts in behavior patterns and predict how it will impact a credit union’s key metrics.

Consumer spending plummeted in the spring, but began to recover over the summer, buoyed by recovery aid and stimulus measures. Grocery spending remains up, but certain categories remain hard-hit by coronavirus fears and economic hardship, such as accommodation, food service, recreational entertainment and transportation.

Now after its short upward trend, the signs of recovery look uncertain again. Millions of Americans are still out of work, but enhanced unemployment, stimulus payments and the small business Paycheck Protection Program have ended. There aren’t many signs of progress toward new aid packages. This, coupled with the possibility of renewed outbreaks as the weather turns colder, leave many heading into the fall and winter with worry and unease.

The key takeaway for credit unions when it comes to spending patterns is that there are many factors influencing behavior and the factors are in continual flux. The situation is highly dynamic. High-level trends indicate the likelihood of continued decreases in cash and check transactions, coupled with ongoing demand for online shopping and contactless payment. But, when it comes to understanding the crucial national, local and individual circumstances driving how and where your members are spending (or not), predictive analytics is critical for tracking and synthesizing the complex situation.

Recognize Strategic ­Opportunities to Increase Income

The combination of credit losses and low interest rates have most credit unions bracing for slimmer earnings. Cost-cutting and creating efficiencies can help – but only so much. Though the coronavirus hardship has been widespread, it’s not a blanket effect. More advanced predictive analysis can reveal the best pockets of untapped opportunity.

Though many credit union leaders may have a sense of where opportunity lies, today’s financial climate doesn’t afford the luxury of following a hunch, only to be wrong. At the same time, leaders don’t need to train themselves on the nuances of clustering, classification-based machine learning techniques or market-basket analysis to form data-driven strategies. Instead, predictive models can be implemented to continuously mine and study data to look for associations, patterns and likely outcomes. These insights make it clear where targeted efforts will be the most rewarded.

When a southeastern U.S. credit union was looking to support its auto loan portfolio after originations plummeted in the spring, it didn’t simply roll out a new promotional offer to the entire membership. First, the credit union employed predictive analytics in a valuable effort to determine which type of auto offer made sense for which members. Using the list of members with a “high likelihood” to act on a loan offer, the credit union was able to create a series of intelligent email-based campaigns using insights from the predictive models.

Traditional notions of “making predictions” evoke speculative guesses and gut feelings – not the sort of things that inspire confidence or make for sound strategy. The field of predictive analytics turns those ideas completely on their head, using the tools of science and technology to help credit unions analytically, accurately and confidently assess what the future may hold.

Karan Bhalla CEO CU Rise Analytics Vienna, Va.

Abhishek Kamodia Lead Data Scientist CU Rise Analytics Vienna, Va.