Founded in 2010, this subprime auto lender exists to give people with less-than-perfect credit a way to get on the road. As the portfolio grew, so did delinquency. Agents were already stretched, and leadership didn't want to solve the problem with more headcount.

• 8% lift in pre-charge-off recovery
• 45% increase in self-serve payments
• 718% increase in digital engagement
• 23% reduction in cost-to-collect

The challenge

Traditional collections meant blasting calls, emails and letters. Consumer behavior was moving the other way: fewer answered calls, more frustration and slower recoveries. The team believed delinquent customers should be treated the way they are when shopping online, with service that is personalized, convenient and on their terms.

The solution

Prodigal's proCollect team unified data across the lender's CRM, dialer, payment portal, email and text platforms. AI and machine learning models then scored every consumer daily on intent to pay, channel affinity and affordability, building a 360° view of each account. With that, the team rolled out:

• Channel strategy: the highest-ROI channel for each consumer, whether call, email or text.
• Timing: outreach when payment likelihood was highest, by time of day and day of week.
• Templates: more than 200 branded SMS and email templates matched to personas, such as first-time delinquents, habitual late payers and consumers in hardship.
• Reactive outreach: real-time messages in response to high-intent consumer actions.

The results

The lender began with a champion-challenger rollout, testing Prodigal on 33% of its pre-charge-off portfolio while the rest stayed on the legacy process.

• +6% payments collected through stronger text and email engagement
• +27% lift in the 30 days past due bucket as models improved with data
• +8% overall payments with 75% of accounts on the new program

Agent productivity rose as well, with a steady increase in dollars collected per agent per hour. The lender has since expanded Prodigal to 100% of pre-charge-off accounts and is preparing to bring the same approach to post-charge-off recoveries.

Download the original case study (PDF)