Smart Prioritisation of Unpaid Debt Collection
The Collections Director manages 80,000 customer files with overdue payments and needs to know who is highly likely to pay after a simple friendly text message versus who requires legal action or an external agency.
Where the money comes from
What to do now
- 1Segment the current portfolio by past collection outcome.
- 2Train the payment probability model by channel.
- 3Assign a channel (text message, call, agency, legal) according to cost and probability.
For your committee
One page with the decision, the impact, the risks, the steps and the KPI to track.
The print dialog opens: choose "Save as PDF".
Watch out: more volume is not always more profit
Pressing everyone equally recovers a bit more in the short term but burns customers who would have paid with a text message. Aggressive collection on a good, normally punctual customer means losing their LTV.
KPI for the committee
Net recovery (collected - collection cost) per euro overdue.
Cash
Accelerates the collection of ~€9M (cash).
Detail for your data team
Proposed technique: Logistic Regression
Why: Logistic Regression delivers a calibrated recovery probability that makes it possible to allocate collection budgets and to justify to auditors why each action is applied.
Cost of getting it wrong: Phoning everyone manually sends operating costs soaring; taking a good but forgetful customer to court destroys the business relationship.
Data needed: Collections CRM: Debt amount, days overdue, previous payment history and payment promises kept.; Contact Data: Response to previous calls, opening of emails/texts and declared employment status.