Case Study: Fraud Detection Automation

Fraud Detection With Automated Analytics

  • For one of the Top 10 international banks, Inspirient automatically scanned high volumes of financial data for suspicious account behavior
  • The Inspirient Automated Analytics Engine can detect account behavior connected to online banking fraud before fraudulent transactions occur and notifies the banks risk team in time
  • The system safeguards the bank through securing their clients trust in online banking while lowering cost connected to fraud
Credit card fraud on the rise
  • 280% year-over-year increase in account takeovers 2016
  • 45% increase in account takeovers between Q1 and Q2 2017
Example: Benford's Law automatically applied to financial data
  • Financial data automatically scanned within minutes
  • Suspicious account behavior and transactions detected
  • Alerted functional team

Innovation and Benefits for the Customer

Every year, $5 billion of damages is caused through account takeovers in the US alone. Besides the financial damage and the high insurance cost connected to fraud, the trust of customers in financial institutions and their online services is threatened. Currently, there are several high-end fraud detection tools but finding suspicious behavior using them takes too long to prevent risky transactions. Even worse, transactions are not detected in time to meet the operational time-frame to re-book fraudulent transactions. Another issue for financial institutions working with these tools is that the high amount of false positives detected implies a very high workload for the employees who need to sort real fraud from the false alerts.

Inspirient developed a fraud detection solution that enables our customers to detect fraudulent behavior before any financial harm is done. The bank is notified about suspicious processes within minutes and can then prevent unauthorized transactions. Inspirient learns during the interaction with the bank’s fraud experts, so the amount of false positives detected is gradually decreasing. This protects customers from financial and reputational damages and lowers operative risk in the long run.

Usage of Automated Analytics

The Inspirient Automated Analytics Engine automatically scans the financial data through its high-speed comprehensive anomaly scan. Within minutes, the system is able to detect known and unknown fraudulent patterns, also custom patterns can be added. To this end, suspicious account behavior is immediately identified and the risk unit of the bank notified in time so unauthorized transactions can be prevented before they occur. Further, the expert knowledge of analysts is fed back into the system through a feedback loop, combining human instinct and experience with machine speed. This enables a constant improvement of the system, as it gets more and more accurate everyday through the feedback it receives.


The case above describes how Inspirient's Automated Analytics Engine for financial fraud detecion and prevention safeguards digitalization of financial services by protecting financial institutions and their customers from fraud. This lowers insurance cost connected to fraud of banks and strengthens the trust of their clients. Also, it allows banks to fully leverage digitization and deliver value to their customers through strengthened protection mechanisms.

If you would like understand how the Inspirient Automated Analytics Engine can be applied to your use case, please do not hesitate to get in touch!

Further reading

The Inspirient Automated Analytics Engine automates the entire data analytics process end-to-end: From the assignment of input data, pattern and outlier detection, automated visualization of patterns, weak points and opportunities to automatic generation of textual explanations and recognition of the underlying relationships and rules. Most other analytics solutions rarely include these textual explanations and observations regarding the underlying data relations, which are both critical to provide a deeper level of analysis and more actionable conclusions.

Copyright © 2021 by Inspirient GmbH. All rights reserved.