Improving Adherence to Debt Repayment Programs

Case Study


American household debt to GDP ratio is 77% – one of the highest in the developed world.

Both consumers and the credit card companies servicing them can benefit from increased financial health and more sustainable debt levels, yet it is not clear what techniques are most effective in shifting consumer behaviors toward this.

About AFS

American Financial Solutions is a non-profit organization located in Seattle, Washington – over the past 15 years, they have assisted over 250,000 Americans to pay off over 42$ Billion USD in debt. Most of their clients fall in the “Mainstreet” market segment. Although AFS’ core offering is a debt management program which consolidates credit card debt, a significant portion of their impact is predicted by their ability to retain clients in the program and help them along the financial literacy dimension.

The Ask

Although the service offered by AFS is strongly valued by clients at the beginning of the program, a variety of reasons cause certain clients to drop out of the program, leaving them in an even more precarious situation than when they first joined. In many cases, this is due to a lack of healthy financial habits, driven by strong biases in client decision making. AFS asked The Decision Lab to perform research on their clients and the market segment more broadly to answer two key questions:

  1. Why do people engage in unsustainable financial behaviors?
  2. What messaging strategy is most effective in nudging them toward better financial lives?

The final goal of this project was to create a messaging strategy that can be used directly with AFS clients (through channels such as websites, phone, email and text message) in order to increase debt repayment rates.


Our team started by performing demographic and psychographic analyses on the market segment targeted by AFS. In addition to analyzing public datasets, our data scientists performed multiple regression analyses on datasets provided by AFS. Our research yielded several very important insights – in fact, we found that although clients differ greatly from each other, the underlying psychological motivators and barriers that yield unsustainable financial habits are quite close to each other.

Our research led to a series of research insights which we then translated into a messaging strategy for the entire AFS team. We worked with them to determine exactly how these insights translate into phone interactions with councilors, marketing collateral, sales materials, check-in text messages, mail warnings, etc. Our predictive model had an 80% accuracy, giving AFS a far more deliberate way of targeting at-risk clients than ever before.


  • Changes will impact over 300K clients in next 10 years, paying off over 50B in debt collectively.
  • Our models explain 80% of dropouts
  • Our solutions can decrease this rate by 50%

Although we cannot reveal dropout rate, we can say that the impact in debt paid off is estimated to be in the billions of dollars.

Read Next