Established in 1828, Citizens Bank is one of the oldest financial institutions in the United States and has undergone numerous transformations over its nearly 200-year history. In 2019, when Michael Ruttledge, Chief Information Officer and Head of Technology Services, joined the company, Citizens Bank launched the next phase of its transformation journey: a major digital acceleration involving a complete overhaul of its technology stack and a large-scale migration to Amazon Web Services (AWS).
With the increasing sophistication of fraud and the growing volume of digital transactions, effective fraud prevention has become essential for maintaining customer trust and minimizing financial losses. Citizens Bank faced new challenges as fraud patterns shifted from primarily card-based fraud to an increase in check and deposit fraud following the COVID-19 pandemic. In response, the bank initiated an 18–20-month transformation of its fraud management systems in 2023, consolidating multiple applications into a comprehensive platform.
“Our fraud ecosystem was fragmented, and all of our legacy applications worked in silos,” said Ajay Punia, Senior Vice President of Enterprise Fraud Applications at Citizens Bank. “Our tech was also outdated.” The bank relied on batch-oriented processes for fraud detection and analysis, which slowed down response times to attacks. Implementing new fraud prevention strategies could take anywhere from days to weeks. To achieve the speed and responsiveness needed for real-time fraud prevention, Citizens Bank wanted to completely reimagine its approach and build new fraud applications from scratch.
With in-house solutions, Citizens Bank could configure applications based on their evolving requirements and provide the highest level of protection for its customers. Specifically, the bank wanted to build a cloud-based solution that could provide near real-time monitoring, seamlessly integrate with multiple fraud detection applications, and offer the flexibility to quickly adapt to new fraud patterns. The bank partnered with MongoDB for this project, building on the success of its enterprise data cache project.
Citizens Bank built a comprehensive fraud management platform on MongoDB Atlas on AWS. Teams from both MongoDB and AWS worked alongside the bank, achieving a smooth implementation. “Our teams were able to quickly adapt to working with document databases, and the collaborative support helped us move efficiently through the development process,” said Punia. “The project demonstrated the value of having strong technical partnerships.”
The migration to MongoDB Atlas marked a shift from the bank’s previous batch-oriented approach, where fraud risk assessments were typically performed at the end of each business day. Now, the bank assesses transactions for fraud risk as they occur in real time, although the exact timing varies by transaction type; for example, ATM and point-of-sale transactions undergo fraud assessment within subseconds of initiation. As transaction volumes grow, the bank utilizes MongoDB Atlas for scalability to help maintain consistent performance in its fraud prevention applications.
The new platform consists of three main applications: monitoring and detection, alerts management, and claims and dispute management. The first application, monitoring and detection, forms the foundation of the fraud prevention system. It scans all transaction types, including checks, deposits, automated clearing house (ACH) transactions, and wire transfers, in near real time. When the monitoring system flags suspicious activity, the platform alerts the fraud operation team to look into the transaction. The alerts management application provides the team with comprehensive data so that they can investigate and determine the optimal course of action. This level of automation streamlines the process and reduces the team’s manual workload.
Ajay Punia, Senior Vice President of Enterprise Fraud Applications, Citizens Bank
With a platform-centric approach to fraud prevention, Citizens Bank can onboard new transaction types and adapt to evolving fraud trends. The bank started with deposits, which was the area identified as having the highest fraud losses. Now, the bank is expanding to other transaction types, including ACH, wire, and real-time payments. Using the new platform, Citizens Bank reduced fraud losses well ahead of industry benchmarks, and reduced the number of false positives flagged by the system.
“When you have more data available about transactions, customer behavior, and ongoing activities, you develop a better predictive mechanism to stop fraud losses,” said Punia. “With automated insights being provided to our teams, they can make better and faster decisions about suspicious activities or when serving customers through the contact center.”
Citizens Bank leverages MongoDB Atlas’ flexible schema and cloud-based capabilities to attain the agility and scalability necessary for highly responsive fraud prevention. “Initially, it used to take us anywhere from two days to several weeks to implement a new strategy,” said Punia. “Leveraging the new home-grown platform, we can launch a new strategy into production in minutes. The new platform leverages MongoDB Atlas on AWS, which taps into the document model, letting us quickly adjust our data structure and fraud detection rules without the constraints of traditional relational databases.”
Looking ahead, Citizens Bank will continue to enhance its fraud prevention capabilities. The bank is actively experimenting with MongoDB Atlas Search to augment its fraud detection algorithms. “Fraud is always moving, and fraudsters are equipped with more tools, technology, and data than ever,” said Punia. “We plan to start integrating with gen AI to enhance our predictive capabilities and stay ahead of emerging fraud patterns. The capabilities of our platform, hosted on AWS and supported by MongoDB Atlas, will continue to support our evolving fraud prevention needs.”
Ajay Punia, Senior Vice President of Enterprise Fraud Applications, Citizens Bank