In this session, panelists explored one of the oldest challenges in collections: distinguishing between consumers who cannot pay versus those who will not pay. Misclassification has significant consequences—chasing those who truly lack resources wastes money, while misidentifying those unwilling to pay can hand out unnecessary discounts.
Dave Aglira, Senior Director at Abraham Gindin, estimated that 30–40% of accounts fall into the “can’t pay” category, based on skip tracing, reports, and agent conversations. He emphasized the importance of direct engagement: “It helps to have that engagement with the consumer to find out what is going on currently in their lives.”
Elizabeth Mejia-Ricart, Manager of Data Science at Second Order Solutions, highlighted the role of modeling. Bureau data, cash flow insights, and engagement signals (such as whether customers answer calls or click emails) can help differentiate willingness versus ability. She noted, however, that willingness to pay remains “more of an art,” requiring human judgment alongside analytics.
Both panelists agreed that the toughest group to identify is the “can but won’t” segment—consumers with the means but no intent to pay. This group frustrates agents and undermines morale, while also challenging predictive models. As Elizabeth explained, shifts in consumer behavior, such as greater reliance on debt settlement or bankruptcy, can erode model accuracy and demand recalibration.
🧠 Key Takeaways:
- Invest in Better Data: Incorporate bureau and cash flow data where possible to strengthen ability‑to‑pay assessments.
- Leverage Engagement Metrics: Track consumer responsiveness across calls, emails, and texts as indicators of willingness to pay.
- Blend Human & AI Approaches: Use AI prompts to support agents, but maintain flexibility for real‑time human judgment.
This webinar underscored that successful collections require a synergy of data science, operational discipline, and empathetic engagement—a balance that helps agencies, lenders, and healthcare providers optimize recovery while protecting reputation.
