Minimising Losses Through Effective Collections Management

February 14, 2019

By Riaz Jassat, Director of Consulting, Qarar

My previous article, part 1 of 3 on Collections, focused on Good Customer Relations. This article comprises part 2, with focus on minimising bad debt losses through an effective collections strategy management.

Collections is the process of obtaining monies owed from customers who have, for whatever reason, fallen outside of their contractual repayment obligations. The aim is to maintain good customer relations while minimising bad debt losses, through effective use of resources.

The above definition of collections encompasses the following key components:

  • Good customer relations: Maintaining balance between assertive  and aggressive collections activity
  • Minimising bad debt losses: Ensuring that collections activity, from a strategy perspective, is designed to minimise bad debt losses
  • Effective use of resources: Optimising all resources (system, human, operational and other) to maximise collections efficiency, productivity and resource availability

Each of these key components of collections comprise individual challenges, approaches and solutions.

Many lenders are focused on growth and increasing balances while minimising cost. This is coupled with pressure on increased provisions due to IFRS9 principles, where lifetime provisions must be taken on high risk accounts. Collection managers need to employ best-in-class principles to cope with the increased volume of accounts, while maintaining existing full time employees and minimising bad debt loses.

These best-in-class principles are designed to ensure the right account is allocated to the right collector at the right time. This level of prioritisation and segmentation requires a deeper understanding of the customer, including past behaviour and predicting future payment patterns. All consumers are not the same and cannot be treated the same way. For this, collection managers are tasked with effectively segmenting customers based on probability of rolling forward in delinquency so that action can be taken on consumers where it is most required.

Various techniques exist on understanding the behaviour patterns of consumers, with the most common being behaviour scoring. Behaviour scoring includes traditional tools of analysing a consumer’s past behaviour, typically on a monthly basis, to predict the likelihood of reaching a specific level of delinquency within a set time period. An example is 3 missed payments over a 12 month window. Using behaviour scoring is an excellent starting point in prioritising accounts as it provides significant uplift to non-predictive based prioritisation, however, there are a few flaws with this approach:

  • Timing: Behaviour scores are calculated once a month, whilst collections activity is conducted and changes on a daily basis
  • Synchronisation: Behaviour scores predict reaching a level of delinquency (typically 3 missed payments) over a period of time (typically 12 months) whereas with collections, the prediction window must be as small as possible
  • Measurement: For predicting performance, behaviour scoring traditionally only assesses transactional behaviour (spend, payments, balance, delinquency, etc.) in its assessment, yet there are many other data sources to be considered

With these considerations, collection managers need to enhance their account segmentation/ prioritisation using:

  • Various data sources
    • Transactional: Payments, delinquency, balances, purchases
    • Collections activity data: Previous times in collections, number of promises made, number of promises kept, number of calls made, number of no answers
    • Demographic data: Employment sector, age, length of employment, region
    • Credit bureau or external data: Length of time exposed to credit, product mix, product performance, new enquiries, affordability (DBR), etc.
  • Increased frequency of assessment to daily/weekly, with a specific focus on collections and payment activity changes
  • Using new techniques such as machine learning algorithms which are capable of consuming this large amount of data in various formats, to predict probability of roll forward

With these enhancements to the data elements assessed, coupled with the latest methods of assessing behaviour, collections managers will be in a position to accurately determine each customer’s risk profile and allocate appropriate action based on the assessed risk. Resources are optimised, customer behaviours are better understood and roll forward balances as well as subsequent bad debt losses are reduced.

For more information on collections, contact r.jassat@qarar.org.

 

Short bio

Riaz is a strategically focused individual with a proven track record of delivering excellence, skilled in credit life cycle management, retail risk, MIS, project management and client engagement. Riaz specialises in Customer Life Cycle Management where he optimises retail lending portfolios profitability through a deep understanding of Consulting, Analytics and Predictive models across the life cycle, cov

Be the first to leave a comment

Share and follow (social media links)

Assigned tags

Collections

Collections strategy

Bad debt loss

Related blog posts

Collections and the Collector