carrot carrot carrot Change Centers x cognizanti collaborators create-folder Data Science Decisive Infrastructure download download edit Email exit Facebook files folders future-of-work global sourcing industry info infographic linkedin location Mass Empowerment Mobile First our-latest-thinking pdf question-mark icon_rss save-article search-article search-folders settings icon_share smart-search Smart Sourcing icon_star Twitter Value Webs Virtual Capital workplace Artboard 1

Please visit the COVID-19 response page for resources and advice on managing through the crisis today and beyond.


Causality AI Predictions Help Maximize Debt Collection Revenue

The Challenge

A large U.S.-based issuer of branded credit cards incurred nearly $1 billion in consumer credit debt every year. Employing thousands of agents to recover debt from consumers in default increased the company’s collections costs to more than $30 million annually. It also led to an agent turnover rate of 40%.

The company needed a more sophisticated predictive technology to improve debt collections, and turned to Cognizant's banking technology expertise for help improving its debt collection strategies and maximizing its debt collection revenue.

Our Approach

We created an automated process to formulate comma-separated values files as usable information. Using a “white-box” artificial intelligence engine, we helped our client fully understand the behavior of consumers who default on their credit card debt and the likelihood of collecting on those debts. 

Adopting a hypothesis as an outcome, our artificial intelligence based causality engine (derived from information theory) can determine variables that are most relevant to the given outcome.

We applied our causality engine to large volumes of monthly data on accounts already in default to identify which of the company’s collections strategies (for example, method of communication and time of day) would best determine the likelihood of repayment. The engine also helped identify a third category of debtor—consumers who will generate revenue if encouraged to pay down their outstanding debt.

Determine Consumer Behavior With Causality AI

Cognizant designed a model to review voluminous data on both slow- and no-paying credit customers. Using our AI-based causality engine helped our client, identify factors that determine consumer payment behavior. It showed the company that focusing its collections agents’ activities on the subgroup of consumers who are more likely to repay their debts would increase their revenue $5 million to $7 million, while also saving a projected $10 million annually. Furthermore, these higher return collections are expected to increase employee commission compensation, which should lead to a significant reduction in agent turnover rates, hiring expenses and training costs.


$10 million

projected call-center savings

Designed a model

to review voluminous data on slow- and no-paying credit customers


factors determinative of consumers’ payment behavior