Empowering Americans to Live Debt-Free Through Behavioral Modeling & Targeted Messaging
0 min read
Around 8 out of every 10 Americans hold some form of consumer debt, with debt levels on the rise. The growing challenge of debt in the U.S. comes from a variety of borrowing types, including long-term investments like mortgages and student loan balances. In fact, the U.S. has a household debt to GDP ratio of 77% — one of the highest in the developed world.
Fortunately, Americans aren’t alone in their debt repayment journey. American Financial Solutions (AFS) is a non-profit organization that offers credit counseling and financial education to those in debt. In the past 20 years, they’ve helped upwards of 450,000 Americans pay off over $9 billion in debt, providing much-needed compassion and support in the process.
The services AFS provide can be life-changing. But habits don’t change overnight, and some of their clients were struggling to stick to their new debt management plans. The same cognitive biases that make it easy to accumulate debt also make it difficult to establish healthier financial behaviors.
Getting the message
AFS approached TDL for help creating a messaging strategy that could boost retention rates in their debt repayment program. We created a targeted messaging strategy as a vector for change, incorporating evidence-based behavioral interventions into different communication touchpoints that would help nudge clients towards success.
Machine learning and client personas
We started out with two key research questions: Why do people engage in unsustainable financial behaviors? And what messaging strategy is most effective in nudging them toward better financial lives?
To find the answers, we took a two-pronged approach: we looked at publicly available datasets and published research on debt management, as well as anonymized data from AFS’s own clients. By looking at multiple sources of information, we could answer our key questions as they related both to the AFS clients and the market segment more broadly.
Our team used machine learning methods like k-means clustering and logistic regression to analyze 100+ characteristics from each person in our dataset. We used these findings to create customer personas, mapped onto the various circumstances and attitudes of AFS’s client base.
The psychological barriers to debt repayment
After creating personas, we tested for variables that predicted eventual dropouts, starting from the earliest moments of the client’s relationship with AFS. Our analysis found that, despite the wide variation between clients, the underlying psychological motivators and barriers that yielded unsustainable financial habits were similar overall. In particular, regression analysis revealed that certain elements of the onboarding process were setting clients down a path towards increased chance of drop-out later on.
Leading the intervention roll-out
After we had zeroed in on the barriers to successful participation, we were able to generate interventions to address them. We worked closely with AFS to develop targeted nudges and incorporate design ideas at specific touchpoints.
These nudges spanned multiple channels, including text messages, sales materials, marketing collateral, and phone interactions with AFS counselors, to name a few. Even minor changes, when enabled early on in the sign-up process, empowered at-risk clients to regain control of their financial situation.
To support AFS in implementing our nudges, we created a schedule for rolling out recommended interventions, and provided continuous support throughout the process. We prioritized open communication between our team and AFS to ensure that feedback from staff could be iteratively incorporated during the rollout.
The most important part of intervention implementation is monitoring. Were our interventions actually boosting success rates? To verify that they were, we provided AFS with a global testing strategy to help measure the total impact of our rollout on dropout rate. The testing strategy highlighted AFS global completion rate (i.e. how many clients paid off their debt) and the average percentage completed prior to drop-out (i.e. how much debt clients paid off before dropping out).
Nudging towards financial freedom
Our global testing strategy found that, using our onboarding nudges, AFS saw a 50% decrease in client drop-out rates. With the span of their work, this resulted in millions more dollars repaid than clients would have seen under the previous onboarding system.
Debt repayment is a widespread challenge in the U.S., affecting the large majority of Americans. We’re proud to support organizations like AFS in their mission, using behavioral interventions to help keep Americans on track to a stronger financial future.