Organisations across sectors are focused on harnessing data to improve the customer experience. Access to customer data has exponentially increased in the last decade, and traditional data analytic processes can’t keep pace with the speed at which this data can be processed, analysed and used to generate insights. 

How machine learning and AI in debt collection is making a difference

CGI is addressing these challenges through the use of artificial intelligence (AI) and machine learning (ML), and this is creating new opportunities to improve the customer experience in collections. With decades of collections experience and data science expertise, we collected data generated from our next-generation credit innovation and management platform, CGI Credit Studio, and conceptualised new ways to improve operational efficiencies, reduce bad net debt, and create a personalised customer experience.

Helping a technology financial services client overcome their collection challenges

This client had been facing numerous challenges, including low customer contact success rates, high dependency on a call centre, operational inefficiencies, regulatory compliance issues, and cost reduction pressure. To help the client manage these challenges, CGI implemented AI and ML models that deliver predictable outcomes for customising default management strategies and reducing operational costs.

The client turned to CGI to integrate our AI/ML models within its existing CGI Collections360 solution suite. These models use first-party user data and our own CGI Collections360 data to learn from customers and predict the best channel and time to communicate with customers.

Implementing AI/ML models to determine best channel and time

We implemented two different AI/ML models for the client that enable it to know what is the best method of contact and time for getting in touch with a particular customer. With its existing default management strategy, it had a high dependency on its call centre as a primary contact channel; 97% of customers were called through that centre, with only 2% receiving emails and 1% receiving texts. With only 3 months of ML data and model training, the client’s new ML-driven strategy reduced its call centre contacts to 85%, while increasing email contacts to 10% and text contacts to 5%.

Overall, the client experienced the following business outcomes:

  • Significantly increased contact rates

  • Reduced operating expenses due to a reduction in call attempts

  • Increased payments

  • Reduced roll rates

CGI’s machine learning automated decision engine

Using the CGI Pulse intelligent process automation platform, CGI is helping collections clients improve the customer experience today, as well as prepare for the future of intelligent collections. Our ML-automated decision engine delivers the end-toend capabilities required to take full advantage of ML and AI to transform the entire collections life cycle. CGI Pulse is designed to increase customer satisfaction and deliver a wide range of other business benefits, including higher customer contact rates, more effective treatment strategies, greater efficiencies, increased debt recovery, and more.

Our innovative ML-based engine delivers the following features:

Customised ML-based predictions

• Deliver pre-built collection models

• Predict customer behavior

• Enable data-driven decisions

• Create tailored customer journeys

Conversational AI

• Automates agent conversations

• Provides human-like intelligent conversations

• Enables omni-channel bi-directional experience

• Enables dynamic workflows

Dynamic rule learner

• Optimises decision trees automatically

• Uses history to drive predictions

• Simplifies business rules


Unsupervised AI

• Captures key metrics

• Mines mountains of data

• Finds micro-segments and patterns

• Delivers insights to drive change quickly

With these capabilities, our collections experts can help you accelerate your ML journey by:

  • Connecting the dots with your customer data, CGI Collections360 data, and third-party data for intelligent journey orchestration
  • Implementing an ML-driven approach for decision-making
  • Personalising treatment strategies to improve collections
  • Harnessing the power of AI in debt collection to reduce costs

Contact us today to learn more about transforming your credit and collections operations with CGI Credit Studio AI/ML capabilities


Download PDF