The Client

A large Financial provider with over $2 Billion in AUM

What The Client Wanted

One of the largest business line of the client was Credit Cards and the client has been facing increasing NPL even after increasing their collection budget substantially

The client wanted to improve its collections process by removing inefficient manual process, making the process real time intelligent and digitally driven The client wanted an automated end to end intelligent platform to manage their debt management and collection process

Challange

The client was having an inhouse Analytics team who have been providing the collection department with various behavioural scoring inputs. As these scores being delivered in the beginning of the month and not dynamic in nature the collection department was facing challenges in making the collection process efficient

Solution

We have deployed CreditNirvana platform in 7 weeks duration which included the Decision Collet, EarlyCollect, RoboCollet and TeleCollect Modules

  1. Adopted the new ML Models in the DecisionCollect working in collaboration with the inhouse analytics department.
  2. Adopted ML and Rule driven collection strategy recommendations
  3. Adopted multi lingual two way ( Chat and Voice) bot driven digital collection services
  4. Adopted self-service digital collection services
  5. Adopted TeleCollect automation for automating call centre allocation process, dynamically calibrate customers payment propensity and execute automatically the next best action
  6. Strengthen customer profiling using both structured and unstructured data on near real time basis
  7. Improve agent performance using speech analytics and intelligent automation

Impact

CreditNirvana has conducted an impact study with client after 6 months of deployment and below is the some of the statistics of this study

  • A 7% increase in funds collected
  • 25% increase in bucket 1 Roll backward
  • 28 % reduction in Bucket X bounce rates in new Credit Card holders
  • 34% reduction in collection expense amounting to US$ 1.1 Million