This article was written for our sponsor, Consumer Education Services Inc.

According to a study by Northwestern Mutual, the average American has around $38,000 in personal debt — exclusive of home mortgages.

As a growing debt load that stems from student loans, credit card bills and medical expenses begins to affect a greater number of Americans each year, some companies have started leveraging technology in an attempt to alleviate the pressure of the repayment process for consumers.

Typically, debt collection efforts include scheduled letters, emails and phone calls from the agency where the debt is owed. If an individual becomes delinquent on their debts, then the original creditor may engage a collection agency for more aggressive attempts to collect the balance. This can understandably be stressful for debtors, as they’re constantly reminded their outstanding payments are accumulating interest, affecting their credit scores, and impacting their financial future.

In many cases, machine learning technology can help alleviate the pressure of consumer debt repayment. By analyzing an individual’s past repayment behavior, AI programs can improve the repayment process for the consumer by developing more effective repayment schedules and recommended payment amounts.

“With machine learning technology, financial counselors can aggregate budget, income, debt load and savings data from a number of consumers and build models that can tell us, for example, whether or not an individual may be more successful using a bi-monthly payment program,” explained Mike Croxson, CEO of non-profit credit counseling agency Consumer Education Services Inc. “These are the kinds of things that machine learning can do to help consumers be more successful.”

Although this specific type of technology is still in the preliminary stages, it’s already changing the way companies and consumers think about the debt repayment landscape. And it’s been in use for longer than you might think.

“I’ve been in this industry for more than 20 years and financial institutions such as banks and investment firms have already been using machine learning,” said Diane Chen, executive director of the Institute of Consumer Money Management. “Using that model, the traditional credit counseling industry can refine their process, which typically has a one-size-fits-all approach to payments. Machine learning can really help us offer a more individualized approach to client communication and repayment schedules.”

Chen’s team at ICMM uses aggregated data from non-profit credit counseling agencies to analyze repayment trends by a variety of demographic categories. This helps build a scoring model to determine a more personalized approach to the debt repayment process.

The model for ICMM’s machine learning initiative took 18 months to build. Once it was finished, the company teamed up with CESI to beta test the technology and refine the model using real-time analysis of client data.

“We’re using this technology in the financial counseling process to help us integrate communication and interventions to help clients be as successful as possible,” said Croxon.

CESI has tested its machine learning initiative multiple times, which has yielded a model capable of generating substantial benefits, as well as more accurate predictive capabilities. The ability to test the accuracy of consumer behavior based on modeled assumptions has been a significant advantage of the collaboration between CESI and ICMM.

Information gleaned from the testing also revealed key information about the target audience for AI-based debt repayment technology.

“We made the presumption that millennials would want to be communicated with by text message, whereas older generations like me would be more interested in traditional methods of communication,” Croxson said. “Ultimately, the data told us we were wrong — everyone wanted more contemporary methods of communication instead of traditional means.”

Currently, the principal data the model collects revolves around payment history, but Chen hopes in the future they’ll be able to analyze things like monthly spending categories, incomes and more to improve the technology’s predictive capabilities.

For the time being, CESI and ICMM hope they’ve positioned themselves to be uniquely ahead of the machine learning debt repayment curve, and they believe this trend will likely become the norm.

“The credit counseling industry is important for consumers’ financial well-being, and that means we need to have an eye toward the technology of the future,” Croxson said. “Somebody in our industry needs to lead the effort. If not us, then who?”

This article was written for our sponsor, Consumer Education Services Inc.