Financial institutions are sitting on vast reservoirs of customer and transaction data—but for many, that data still behaves more like scattered archives than a strategic asset. As banks look to compete on personalization and speed, data strategy is shifting from a back-office IT concern to a core driver of business value.
They have learned to reduce operational silos so they can share data more effectively, creating a more complete and actionable view of customer behavior. While many banks have taken advantage of these new capabilities, others still have not, or are only analyzing a small subset of the available data.
In the Debit Payment Data: A Business Strategy, Not Just an Initiative report, Ben Danner, Senior Analyst of Debit at Javelin Strategy & Research, examines how financial institutions can maximize the enormous amounts of customer and transaction data they hold.
“Large institutions are already making moves to bring customer data into their apps through services like Plaid and budgeting and forecasting tools, which used to be only in the domain of third parties,” Danner said. “Mid-size banks and community banks cannot wait with data strategy.”
A Wealth of Information
Product managers for offerings like debit and credit cards are familiar with key performance metrics such as the percentage of deposit accounts that have a debit card attached, how many of those debit card holders are actively using the card, average ticket size, and so forth. These are important transaction data points that anyone in product management should tracking for card products, even though each bank may analyst its data somewhat differently. However, the best banks are moving beyond these basic metrics.
“That’s a good first step,” Danner said, “But if you’re really looking to understand the market better, you’re going to have to go deeper than just looking at basic performance metrics on your card program.”
Deeper segmentation is where product managers and analyst teams can really excel, drilling down into areas like merchant categories, transaction types, and ATM usage. Every bank is sitting on a gold mine of transaction data that describes the habits of its customers. Historically, the challenge has been that banks haven’t had the right tools or enough time to dig deeply and make sense of all that data.
“You can have terabytes of servers worth of processing data, but it’s meaningless until you start digging in and doing the analytical work and interpretation of it,” Danner said. “This isn’t mind-blowing stuff, but there are important segmentations to look at, like breaking down your spend types.”
Challenges for Smaller Banks
Large institutions are already integrating artificial intelligence agents directly into their analytics and intelligence platforms. That leaves smaller banks needing to accelerate their modernization efforts to remain competitive.
Many of these smaller banks simply don’t have the resources or analytical tools to fully exploit this data. While organizations like Chase or U.S. Bank can dedicate teams to these initiatives, mid-tier and smaller banks don’t necessarily have an easy way to achieve the same outcomes. A smaller bank might have one product manager overseeing multiple offerings, or a single card lead responsible for the entire debit and credit card portfolio.
They often become the primary interaction point between the bank and third-party providers working on card programs. Fortunately, there are several ways they can mitigate these constraints by partnering with analytics and consulting firms.
“A lot of banks sit on terabytes of data like this, but they don’t have the people to analyze it,” Danner said. “The way is through partnerships—working with research firms, and working with your processor or your networks. Visa and Mastercard have consulting analytics services that they can offer to do all of this. They can do that kind of analysis, so it’s not all on you.”
The Promise of AI
Discovering patterns and actionable insights in transaction data is one of the areas where artificial intelligence excels. Many fintech vendors, particularly Fiserv and FIS, are launching agentic tools that can perform this kind of advanced data analysis.
The larger the bank, the more likely it is to build its own AI models in-house, as it has the necessary resources and teams. These institutions can develop and deploy agents internally for a wide range of use cases. Moving down the scale to smaller banks, they are more likely to partner with their processors or vendors to leverage these capabilities.
“If your core or your bank is working in partnership with Fiserv, you can go into their menu and pick out different products and services you want in this marketplace dynamic,” Danner said. “They have agents that you can pick out, and you pay individually for these kind of services.”
“Fiserv has already started building these agents, including ones for risk management, regulatory compliance and reconciliation,” he said. “FIS has a financial crimes agent to be used for AML and decisioning. They can deploy these agents that pull data from multiple silos together to create more of a holistic picture of what’s happening.”
Breaking Down the Silos
One of the key challenges in fully leveraging this data is breaking down organizational silos—bringing information from across the enterprise together for integrated analysis rather than keeping it fragmented.
Maximizing a bank’s data means combining multiple types of information and generating innovative insights, such as transaction frequency patterns. These populations can then be used to build a business case for new products.
“The problem was we have a lot of data, but not a lot of folks necessarily to interpret all of it,” said Danner. “Now we have agents, so maybe the agents will interpret it in a nutshell.”
“Banks are these data gold mines,” he said. “There’s a lot more that you can use instead of sitting on the basic performance metrics. But even more so, how you analyze and interpret that data is really the key to success.”
