Digging for golden evidence

Data mining: a tool for all fraud examiners

By Tammi Johnson, CFE and Judd Hesselroth, CIA, CISA

Fraud examiners need only basic computer skills and the help of systems administrators to mine for gold and find vital evidence for fraud examinations.  

A routine accounts payable audit uncovered a split purchase for expense office furniture made with a travel and entertainment credit card. Further review indicated that Sam, a veteran employee, had actually been making a lot of expensive purchases through a local supplier through several methods even though the company had a preferred national furniture vendor. Monica, a Certified Fraud Examiner, had strong suspicions that Sam may be involved in procurement fraud but lacked conclusive evidence.

To the uninitiated, a simple split purchase made with an improper card, or an invoice discrepancy might not have much meaning, but to a CFE skilled in data mining, every nook and cranny might be a treasure trove of valuable evidence. Even though she had only basic computing skills, Monica was able, with the help of her systems administrator, to search and analyze her firm's databases and discover the incriminating material she needed. Like Monica, you can put on your mining helmet and dig for gold. 1  

Business or nonprofit databases can be minimal in size or grow to several terabytes. Any of these data sources may hold hidden yet vital information you may be seeking during a fraud examination. No matter your background you can become skilled at data mining. In fact, it''s no longer an optional skill set.

Data mining has also been called database reporting, data analysis, and other similar names but generally can be defined as "the automated extraction of hidden predictive information from databases." The tools discussed here will assist with automated extraction.

With the explosion of databases that track customers, sales, and other information, data mining is now being used to increase revenues - through improved targeted marketing - and to reduce costs - through detecting and preventing waste and fraud.

As a fraud examiner beginning any investigation, you always have some pre-work that includes deciding what data that you want to mine, for what period of time, and from what system. Before you use the actual mining tools, you need to identify the problem to be solved; identify, collect and prepare the right data; interpret the data; and monitor the results. The real key to success, however, is having a thorough understanding of your data.

For full access to story, members may sign in here.

Not a member? Click here to Join Now. Or Click here to sign up for a FREE TRIAL.