Case Study of FINNABLE

Turning Repayment Signals Into Earlier, Smarter Collections.

Finnable partnered with Fundamento to automate early-stage personal-loan collections across Voice, IVR and WhatsApp, using cohort-specific journeys to engage borrowers before missed payments became deeper portfolio risk.

Written By

Team Fundamento

Published

Managing Repayment At The Moment It Matters Most

Finnable is a digital lending business focused on personal loans.

Fundamento’s deployment focused on the earliest stage of post-disbursal repayment, where timely engagement can help borrowers stay on track before repayment behaviour deteriorates.

Early Repayment Requires Different Strategies For Different Borrowers.

Not every borrower enters repayment with the same risk or intent.

Finnable needed to distinguish borrowers who were likely to pay from those requiring stronger intervention, while maintaining high-volume, multilingual engagement across repayment cohorts

Cohort Journey
FEMI First EMI repayment
Bounce & Pay Borrowers requiring post-bounce engagement
PDM IVR Pre-due / repayment reminder
Mandate Bounce IVR Mandate-failure recovery
Bounce Post-bounce recovery
Where Leakage Was Concentrated PDM / FEMI: stronger repayment performance Bounce: materially lower payment rates This made cohort-specific intervention more important than a single collections script.

Signal OS™ Turned Repayment Behaviour Into Actionable Cohorts.

The deployment used repayment status, mandate behaviour and cohort position to determine which journey should engage the borrower and when.

Signal What It Revealed
Repayment Status Whether payment had been made
Mandate Status Risk of payment failure
Cohort FEMI, PDM, Bounce or Mandate Bounce
Engagement History Prior interaction and response
Journey Position Where intervention was required
Architecture Layer Deployment Capability
Customer Signals Repayment status • Mandate status • Cohort
Decision Intelligence Cohort-specific prioritisation
Journey Intelligence PDM • FEMI • Bounce • Mandate Bounce
Execution Layer Voice AI • IVR • WhatsApp
Human Intelligence Automated deployment; no human calling layer
Repayment Signal → Cohort Identified → Journey Selected → Timed Outreach → Payment / Further Intervention → Learning

One Collections Strategy Would Not Fit Every Repayment Cohort.

Journey Objective After Fundamento Outcome
FEMI Encourage first EMI payment 90–91% paid
PDM IVR Pre-due repayment engagement 70–77% paid
Bounce & Pay Recover post-bounce payments 17–21% paid
Mandate Bounce IVR Recover mandate failures 9–11% paid
Bounce Recover delayed payments 17–21% paid
Cohort Identified → Appropriate Script → Voice / IVR / WhatsApp → Payment Behaviour Captured → Next Intervention

Collections Agents That Adapt To Repayment Context.

Agent / Layer Responsibility
Voice Agent Cohort-specific borrower engagement
IVR Structured repayment outreach
WhatsApp Supporting reminders / payment nudges
Workflow Engine Calendar- and cohort-based orchestration
Language Layer Collections across 12 languages

A Fully Automated Repayment Layer.

This deployment was designed for autonomous Bucket-0 repayment engagement, with no human calling layer documented in the source.

Automated Layer Role
Cohort selection Determines borrower treatment
Voice / IVR Executes repayment conversation
WhatsApp Supporting payment engagement
Journey logic Determines next action

Capabilities Are Not Projected. They Are Measured.

90–91%

First EMI Payment — FEMI

70–77%

PDM IVR Payment

17–21%

Bounce Payment

9–11%

Mandate Bounce IVR Payment

12

Collections Languages

From Collections Automation To Repayment Intelligence.

  • First-EMI behaviour patterns
  • Mandate-failure signals
  • Bounce recovery cohorts
  • Channel performance intelligence
  • Multilingual collections learning
Started With Expanded Into
Bucket-0 collections Cohort-specific repayment journeys
Repayment reminders Behaviour-led intervention
Voice outreach Voice + IVR + WhatsApp
Fixed repayment logic Cohort intelligence

Related Articles