How one company used data visualization to stop bank fraud.
For an online bank, every day means collecting massive amounts of data. From tracking online transactions to money flows, they gathered important insights for more than just their every day, but to help detect fraudulent charges as well. But with so much data, how do you sort it all without getting siloed?
Using relational technology and data visualization, the team was able to flag customers, transactions or behaviors, that didn’t line up. But the biggest help to the team came with data visualization.
The investigation team improved their capabilities to find fraud with graph analysis and visualization layers. The result?
FINDING THE FRAUD
Being completely online, BforBank’s systems collect massive amounts of data every day. From tracking online transactions to money flows, they gather important insights for more than just their every day, but to help detect fraudulent charges as well. Monitoring this data is imperative to lowering risk and financial losses.
The Issue and Data at Hand
Using relational technology and data visualization, the team was able to flag customers, transactions or behaviors, that didn’t line up and investigate. Querying connections within the data to confirm fraudulent activities was often a long and tedious process though, sometimes taking several hours to make connections.
With so many complex solutions , the cases required access to information buzzying down data silos causing additional technical resources and slowing the process down. AKA Huge potential larger losses
But the biggest help to the team came with data visualization.
So, with so much data, what’s the solution? Network visualization and analysis!
By implementing a graph analysis and visualization, the risk + compliance unit and the IT department was able to improve fraud detection and reduce investigative time.
Designing a data model that compiled customer data, bank transfer orders, check cashing activities, and IP addresses into a graph, an extensive overview of their data was created. Once quickly transferred to a visual realm, the team was able to spot hidden connections of suspicious clients by reducing blind spots – they could see EVERY connection!
What was once a complex system of finding relationships was now a visual playground for their data.
BforBank finally had the ability to easily query connections and specific patterns within the data thus detecting new fraud situations FASTER.