Leading UK Retail Bank

As more internet banking services become available, the emphasis on visiting the branch decreases due to the convenience of carrying out routine tasks online. As the breadth of products and services continue to increase, so does the issue of security in the financial industry as the associated make it harder to protect a payment system. Fraudsters are increasingly sophisticated and successful, especially as commerce and financial transactions move online. Here, it’s easier for fraudsters to evade detection, use stolen credit card information, impersonate individuals and take over online accounts.



Whilst organizations have turned to existing anti-fraud solutions, many don’t have the capability to see every type of fraud, so potential red flags fall through the gaps. The good news is, as business moves online, the evidence of internal or external fraud often lie in the massive amounts of unstructured machine data, commonly log files, generated within business applications, IT infrastructure and security systems. This makes fraud detection and prevention a big data challenge that organizations can use to implement process and actions based on fraudulent activity. So how can businesses leverage big data within their own organizations? Let’s follow the story of a leading UK retail bank.

With 14 million+ active customers and multiple brands, the risk of fraud activity had greatly increased and recently, due to geo-political tensions, the bank wanted the ability to monitor unusual behaviour and fraudulent activity mainly from Non-UK Countries. In particular, this bank needed to gain insights into payments-related fraudulent activity, including payments going to a single account from multiple users & credit card applications and approvals.



By having additional monitoring in place, the bank was confident it would be able to implement a process and associated actions that would protect customers and enterprise information, assets, accounts and transactions through the real-time, near-real-time or batch analysis of activities by users and other defined entities. Ultimately, they wanted a solution that automated the detection of potentially fraudulent activity and flagged that activity for review. The bank knew exactly what it wanted and the importance of getting it done, but they were struggling to put the necessary tools in place.

Initially, the Bank’s Digital Security & Operational teams had no means of monitoring these fraudulent activities apart from relying on Cyber Security team to provide a view of that data. GlobalLogic was on-boarded to provide professional services to the bank. With the adoption of Splunk’s Machine Learning Toolkit, we began indexing relevant machine data before searching and correlating it to identify the patterns of fraud. Doing so enabled us to put alerts in place that flagged fraud attempts in real time and prevented them from impacting the bottom line.



Our AI-enabled approach and engineering capabilities enabled us to analyse data coming from multiple sources, such as F5 devices, authentication systems, transaction processing systems, payment and billing systems, databases etc. Leveraging all the associated data helped to detect anomalous internal and external behaviour, as well as indicators of failures through statistical analysis and machine learning capabilities.



All these insights were collated into a customised form-based dashboard, with drill downs providing easy access to targeted data for their investigative needs. We also created rules & dedicated dashboards capable of correlating possible fraud indicators across all channels. Not only did this eliminate silos and manually intensive and cumbersome investigation processes, the bank now has a 360-degree view of their data. Digital Teams can see all customer activity in one place and look for anomalous changes in patterns in single or multiple channels that might indicate fraudulent activity.