THE BUSINESS APPLICATION
Where and how to use it
Use it in collections prioritisation and restructuring review, alongside treatment eligibility, customer circumstances and operational judgement. It can support a more explicit discussion of whether a recovery is likely to persist.
A borrower clears arrears after a revised payment arrangement. The team tracks whether the account remains current through the agreed follow-up period before describing the intervention as a sustained cure.
From evidence to a decision
- 01Arrears episodeDefine when distress begins
- 02Possible cureReturn to agreed performance
- 03Follow-up periodWatch for a renewed default
- 04Sustained recoveryCompare cure and relapse risk
DATA REQUIREMENTS
What records does it need?
These are the records your team would bring together for this analysis. The exact fields and history needed depend on your lending products, the question you want to answer and the period you want to assess.
| Record category | What it contains | Why the detail matters |
|---|---|---|
| Distress episodes | Account ID, episode start, severity and relevant prior behaviour. | Repeated episodes must be distinguishable. |
| Treatments and receipts | Dated arrangements, restructures, contact/action records where permitted, and payments. | Retain timing and eligibility context. |
| Cure and redefault events | Versioned definitions, event dates and observation end. | Open episodes and incomplete follow-up remain explicit. |
Past loan outcomes help assess how well an estimate reflects your borrowers. For a new decision, use only the information available at that time; later repayments help you review the result afterwards.
Understand data readiness →WHAT YOU RECEIVE
The output
Horizon-specific cure and redefault estimates with the assumptions needed to interpret sustained recovery.
WHAT TO WATCH
The limitations
Historical treatment choices are not random. A higher predicted cure probability is not proof that a particular action is effective or appropriate.
FOR RISK & ANALYTICAL SPECIALISTSHow the analysis works+
The modelling approach
Separate classification or event-time components model cure and subsequent redefault. The clock starts at a defined distress or cure event. Treatment assignment can bias observed outcomes, so prediction alone cannot prove which collection action causes a better result.
What your risk team should review
Time at risk, repeat episodes, censoring, treatment selection and consistent cure definitions.
The right approach depends on your portfolio and available history. Review the fit to your borrowers, the reliability of the estimates and the effect of missing information before using the result in a lending decision.