Business Analytics 101: Enterprise Fraud Management


Recently I was listening to an NPR (National Public Radio) documentary about the history of cancer and medicine’s evolution in understanding and treating cancer. This was an amazing story. I would venture a guess that nearly everyone who is reading this post has been affected by cancer in one way or another. So you can imagine my surprise and amazement when I later read a fraud article about cancer. An elderly woman was faking cancer to play on the sympathies of people and cheat them out of their money. This is what makes me so passionate about sharing the importance of fighting fraud and beating fraudsters at their game. Fraudsters, like cancer, have many faces and are often in and out before you know it.

The cancer scam starts small as an individual crime against an empathetic donor. However, once successful, this fraud expands and is characteristic of fraud that costs organizations millions each year. Financial services firms, health care, insurance – fraud is a cross-industry pain. Gartner, a highly respected research firm, defines enterprise fraud and the solutions to mitigate fraud in their most recent report, Enterprise Fraud and Misuse Management Solutions: 2010 Critical Capabilities.

“Gartner defines EFM as software that supports the detection, analysis and management of fraud across users, accounts, products, processes and channels. It monitors and analyzes user activity and behavior at the application level, as opposed to the system, database or network level, and watches what transpires inside and across accounts using any channel available to a user.”

Business Analytics in the fight against fraud

In a recent report by the Fraud Management Institute, business analytics approaches that should and are being used to execute on an enterprise fraud strategy include data integration, fraud detection models, alert management and result evaluation. An effective enterprise fraud data warehouse needs to capture and integrate data from a wide variety of sources and aggregate the data related to all transactions so they are useful for real-time risk scoring.

For example, the State of Washington has investigators doing research on 12 to 25 different technology systems to get a more comprehensive view of a particular employer. Firms need to consider multiple analytical approaches to modeling fraud that extend beyond business rules and include anomaly detection, predictive models and social network analysis. The choice of which methods to use often depends on the particulars of the application and the institution. In general, there is a trend away from the use of business rules as the lone method for defining alerts.

Organizations are moving toward more responsive fraud alerts and a higher level of alert integration across business lines and various geographical regions. In some cases, the alerts are integrated into a single case management tool (see the SAS white paper Enterprise Case Management) . Finally, organizations should be evaluating results with key metrics that are agreed upon, documented and reviewed regularly. Key metrics can include historical expected losses per account, number of false positives and total exposure. The State of Washington has already estimated an 8 to 1 return on their investment in a workers compensation fraud prevention solution from SAS.

Time to Take a Fresh New Approach

As banking evolves into new business channels, these channels can pose new risks. The first concern is to know and authenticate customers so you know with whom you’re doing business. This is easier said than done, which is why a layered defense should be used. Analytics are critical to that defense. Use technology that can learn from complex data patterns, and use sophisticated decision models to better manage false positives. To learn more, read “Building a Better Banking World.”

Employing a layered hybrid approach for fraud detection helps the client by providing the ability to tailor approaches to specific clients. Included with this hybrid approach are canned analytics that are an important part of ensuring rapid ROI and deployment simplicity.


Many financial institution fraud experts know they need to do more, but as many companies have already poured money into simple solutions it is difficult to explain to executives where the institution stands and what it needs to do next. Many companies can’t even measure how successful their efforts are with the tools they already deploy. Use this simple five-level assessment framework to see get the conversation started.

The business case for using multiple analytical approaches across all organizational transactions will not only give you better monitoring of fraudulent activities, but more accurate behavior profiles that result in incremental detection and reduced false positive rates. This will keep your customers safe from financial harm and protect your financial institution’s reputation.

Closing Thoughts

I hope I have left you feeling more confident in the important role that business analytics plays in staying ahead of fraudsters and keeping them benign. Remember our lady who was running the cancer scam? With sophisticated analytics at work, the day-to-day including doctors’ visits, medicine refills and donation transactions could have been tracked and flagged as suspicious before many of the donation checks were cashed!

When these tracking measures are applied continuously, detective work becomes preventative.

Watch this short “YouTube” video to see how one of our key global customers is working to build a better banking world with less fraud, bigger profits, and more satisfied customers.


About Author

Ellen Joyner-Roberson, CFE

Global Marketing Advisor

Ellen Joyner-Roberson, CFE, is Global Marketing Advisor at SAS where she defines industry strategy and messaging for the global fraud and security markets in banking, insurance, health care and government. With more than thirty years of experience in information technology, she helps clients capitalize on the power of analytics to combat fraud and keep the public safe. This includes bringing greater awareness of how to apply machine learning and AI to detect evolving fraud tactics, while realizing ROI in technology investments. In addition, she consults with clients to reduce fraud losses and mitigate risk across their enterprise. Joyner-Roberson graduated from Sweet Brier College with a degree in Math and Computer Science. Most recently, Ellen has brought to market our Intelligence and Law Enforcement solution called SAS® Intelligence and Investigation Management and a cross industry solution focused on procurement integrity.

Comments are closed.

Back to Top