Case studies / Banking · Collections

How a bank’s loan servicer raised its monthly collection rate by 81%

A tier-1 Spanish bank’s servicer worked every defaulted loan the same way. Two models now decide which loans to work, and through which channel.

  • +81%monthly collection rate
  • −41%cost to collect
  • €4bn+portfolio under management

The situation

The servicer manages more than €4bn in non-performing loans and repossessed assets for a tier-1 Spanish bank. Loans were worked through collection agencies, digital channels and in-house teams, and the quality of the loan data was very low.

The problem

The goal was to recover more with the same team. That means knowing, loan by loan, how likely a payment is and which channel is most likely to get it. Until then, every defaulted loan was worked the same way.

What we built

  1. Payment anticipation

    A causal model estimates the probability of payment for each loan and what drives it, built to cope with incomplete data.

  2. Channel allocation

    An engine sorts every loan between agencies, digital channels and the in-house team, according to its expected return.

  3. More intensity, same team

    Managers spend their time on the loans where a person changes the outcome. The rest go to cheaper channels.

The results

Monthly collection rate
Before0.62%
After1.12%
Cost to collect
Before
After−41%
  • Monthly collection rate up 81%, from 0.62% to 1.12%.
  • Cost to collect down 41%.
  • Assets under the models: €3.4bn with payment anticipation, €721M with anticipation and channel allocation.

Client anonymised. Full case file and reference available under NDA.

Tell us the problem.

Start with a free Value Map: three hours with your team, and a fixed-price plan for your first system five days later.