The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

Using The Power Of Prediction To Pro-Actively Action Intervention In Your Call Centre

For collection operations and risk alignment, a critical success factor is the ability to predict month-end results accurately and at an early stage of the billing cycle.

This has always been a much-contested topic, due to the nature of accurate predicting early in a billing cycle.

The norm of predicting results in operations has always been a combination of tracking performance points to last-month-same-time, last-3-months-same-time and last-year-same-time, but also to take seasonality and unexpected anomalies into consideration.

While using data is a good start, basic analysis is often unreliable and insufficient to accurately forecast performance. It creates inter-month inaccuracies resulting in an inaccurate month-end expectation. Standard comparisons won’t give you the answers you need to create a strategy that will see the desired results, and it often doesn’t identify areas to implement pro-active remedial action and when to implement these actions to achieve your desired result.

The next question would be What does? Is there an alternative?

And yes, there is an alternative: big data analytics with machine learning capability.

Machine learning driven forecasts

By using predictive analytics and machine learning methods, it is possible to forecast results accurately, and predict which areas need remedial actions in the short or long term to have an impact if you are not tracking favourably to your month-end expectations. You can use these insights to inform your business and collections strategy very effectively.

Tools that offer this are straightforward to use and easy to understand. Based on your call centres past performance, machine learning can predict what your results will be in the current month and identify problematic areas.

By graphically plotting metrics and predictions against targets, it's also effortless to understand the data at a glance.

How are operational problem areas identified?

By accurately predicting month end results (in contrast to month-to-date results) early in the billing cycle, you can identify overall problems if collections will come in below target.

If performance predictions indicate overall unfavourable month-end results, a more granular inspection should be done, and individual portfolios/areas evaluated in the same way as overall performance. This will help identify which portfolios/area are under pressure, and require prioritised focus. This granular breakdown will also give you an overview of the most and least successful portfolios, and help you prioritise your workforce accordingly.

It is also essential to identify which performance metric (connects, Right-Party-Connects, Promise-to-Pay etc.) can be linked to the predicted unfavourable performance. Some performance metric intervention requires "back-office" focus as opposed to call centre agent performance focus (matters worked, negotiation etc.).

By breaking your targets down to a granular level, you’ll be able to identify both portfolio and performance metrics that are in need of pro-active remedial action. (Click to Tweet!)

How will I know whether the remedial action I’ve taken is effective?

The benefit of machine learning and accurate forecasting is that these predictions and problem area identification will take place very early in the billing cycle (or reporting period), allowing for sufficient time for any remedial action to make an impact. And if you continually feed your machine learning algorithm with updated information (data), your predictions will update dynamically, and increase in strength.

If you’re interested in using data analytics to optimise your collection performance, download our expert guide. You can also read more on our predictions tool, Prosperity.

increase your collection and recovery yields with data analytics

Perry de Jager
Perry de Jager
Perry has been involved in Collections and Recoveries for the past 12 years, spending time in different market segments ranging from law firms to investment companies. At Principa, Perry has worked on extended projects within both South Africa and the Middle East with some of the largest financial organisation, providing on-site consulting within the collections and recoveries space covering strategy, process, people and technology.

Latest Posts

Will Your Machine Learning Models Pass The William Tell Test?

Machine learning models can be used very successfully in many different contexts to predict outcomes for different use cases accurately. These predictions can be used within the business to make better decisions or to operate more efficiently (or both) and can give you an edge over your competitors. Predictive models all follow the same recipe – i.e. train a model on historical data and then apply this model to unseen data to get predictions. If your model generalises well, you have a prediction that you can trust and use to decide "do this, not that" with some degree of accuracy.

Everything You Need To Know About MarketWise: Your FAQs Answered

What is MarketWise? MarketWise is an audience of online users within South Africa which can be segmented and arranged according to a brand or advertisers target market. Online ads can then be served to this audience as they browse the internet.