By Scott Patterson & Amy Logan
More than 40 percent of frauds are initially detected by tips, according to the ACFE’s 2010 Report to the Nations on Occupational Fraud and Abuse. Peter Millar, Director of Technology Application for ACL Services Ltd., points out that the human element of this detection method still requires a person to come forward with suspicions – often after a length of time has passed and the damage has been done.
Data analysis, on the other hand, “is more of a proactive approach that seeks out indicators of fraudulent activity in an organization's data,” Millar said. “It can highlight indicators of fraud soon after they first appear in the data, and not 18 months after the fact.”
Tom Huber, assistant vice president for RLI Insurance Company in Denville, N.J., said he was attending Millar’s session, “The Best of Crimes, the Worst of Crimes: Fraud Stories that Prove the Truth is in the Transactions,” to find out what questions he needed to ask his potential insures to see if they were taking the proper fraud-prevention precautions.
“They call me for insurance so if there’s a loss, it’s covered by insurance,” he said. “My purpose in being here is to find out if they’re doing it right. I’ll say, ‘Hey, you could be doing this, do you know that?’ And if they’re doing it right, then it could be a good insurance risk.”
But if they’re not, he said, he won’t write them as an insured.
Millar said while there is a broad range of analytical capabilities out there, not all business leaders use them or are even aware of them.
“Some organizations have done a great job in applying technology in their fight against fraud while others have a way to go yet,” Millar said. “It's my belief that data analysis can always be used to greater advantage, regardless of the organization.”
However, Kathryn Weatherby, CFE, Fraud Exam Specialist for the FDIC, said that while she agreed with Millar that companies should be using data analysis tools, the reality was that many won’t because it’s expensive to implement.
“In my world, I don’t see the dollars being spent on that sort of detective software,” she said. “I’m hoping that as it becomes required, either by regulations or if the losses get so high that they get bit and it hurts, then they’re going to change. But until then, nobody’s going to data mine those transactions.”
Millar said ACL has a saying: “The truth is in the transactions.” He explained that data analysis enables organizations to look for red flags that might indicate fraud by examining all available data populations.
“It allows broader and deeper coverage than what could otherwise be provided,” Millar said. “It allows for the early detection of fraudulent activity so it can be stopped before it harms an organization financially and in terms of its reputation.
“Data analysis can access, analyze, compare and contrast data from dissimilar systems to expose fraud schemes that would otherwise be impossible or difficult to identify: i.e. Comparing vendor addresses against employee addresses, or vendor names and addresses against barred lists etc.,” Millar said.
But while the technology is amazing and would make it much easier for companies to detect fraud early, Weatherby said she’s skeptical that enough companies are ready to embrace it.
“To a large extent, if you’re into data mining then you’re an IT geek,” she said. “We need to change that perception and we need to stress that with all the automated processing we have today, there is no way you can manually audit for this stuff. We can’t mandate the detection systems either, but when a fraud is identified, that’s usually when these kind of vendors are going to get an appointment in the fraud department. That’s the reality.”
Weatherby said one thing she found particularly useful in the session was that Millar offered a compromise she could share with banks and other subsidiaries that wouldn’t spare the expense to purchase proper data analysis software.
“One of the nice things about this session was that [Millar] showed us how to go through the risk assessment and figure out where the highest frauds are, then figure out what you can automate and just let the automation go so you can focus on other areas,” she said. “It’s still a hard sell, but I was encouraged by the comments and I’m interested in the queries myself.”
The Association of Certified Fraud Examiners assumes sole copyright of any article published on www.fraudconference.com or www.ACFE.com. ACFE follows a policy of exclusive publication. Permission of the publisher is required before an article can be copied or reproduced.