In an era of data-driven loan decisioning, bankers are working to better integrate and utilize their existing customer data more effectively, particularly for their business banking customers. The challenge for many lies in organizing large volumes of data from multiple sources, often stored in disparate systems, that prevent banks from having a truly comprehensive view of their customers.
Despite significant investments in data stores and extraction, transformation and loading (ETL) processing, this data fragmentation poses considerable challenges that lead to inefficiencies in operations, customer onboarding and risk management. More importantly, this lack of integration across data sources increases the bank’s potential exposure to fraud.
Historically, FIs have turned to data orchestration solutions to manage the influx of data from various sources. These platforms typically focus on optimizing data costs and identifying hard-rule declines to save on human processing time. While these solutions provide some value, they often fall short in key areas.
One major limitation is the lack of integration with historical enterprise data. Traditional platforms are generally designed to manage external data sources but fail to incorporate the institution’s own historical data. This omission means that banks cannot fully leverage their past experiences — whether positive or negative — in their decision-making processes.
Another notable flaw is the absence of consistent data schemas that are essential in supporting today’s AI modeling systems. Without a standardized approach to data management, banks face difficulties in extracting meaningful, actionable insights from their data, further increasing ETL expenses. These limitations underscore the need for more advanced solutions that can address the complexities of modern data management.
Identifying fraud through data analytics
Enhanced data analytics often reveal to bankers that small business loans written off as losses typically share common data attributes with other bad loans. This is not limited to obvious data elements, like business EINs or owner Social Security numbers, but also things like secondary contact telephone numbers, nearby street addresses, IP addresses and/or device IDs. Had these loan application-related data elements been readily accessible to underwriters from the start, they could have identified and declined these risky (or fraudulent) loans at the point of application.
Here’s a common fraudulent scenario that a bank might encounter: A business unfamiliar to the bank applies for financing. Identity validations for the business and the business owner initially look good, but there are several related applications based on matching device IDs and bank account numbers coming from totally different businesses, which raises a red flag. With advanced data analytics, the bank could then see that two of these related applications were previously identified as fraudulent and declined. From here, the bank can set up a rule within the automated platform to route applications with this related application fraud pattern to manual review or it can have the system simply decline future applications outright. One of the key benefits of these platforms is their ability to generate AI-driven insights. Through the use of pattern-matching algorithms and machine learning models, bankers can identify relationships and trends within their data that would otherwise go unnoticed.
In today’s lending market, business banking customers expect the same level of speed and convenience as consumers enjoy. By integrating historical data, third-party insights and real-time monitoring, modern data platforms are enabling bankers to make more informed, data-driven decisions in a way that meets customer service expectations, effectively reducing friction for legitimate customers while quickly identifying and rejecting fraudulent applications.
For commercial and business loan portfolio management, these data insights allow bankers to optimize their risk management strategies. By analyzing historical data alongside real-time information, banks can better understand the performance of their portfolios, identify emerging risks and take proactive measures to mitigate them, ultimately resulting in healthier portfolios and improved financial outcomes over time.
In an increasingly complex and data-driven financial landscape, the ability to effectively manage and leverage actionable data is critical for success. By transforming fragmented data into a strategic asset, modern business lending platforms are not only improving operational efficiency but also driving revenue growth for financial institutions.
Will Tumulty is CEO of Rapid Finance, which provides working capital to small and mid-sized businesses in the United States and enterprise solutions to enable lenders to serve small business borrowers.