Insights into behavior patterns and fraud detection

Five insights to consider

1. Companies often wait too long
   Complacency is the enemy of effective fraud-fighting. Organizations that sail along doing business without encountering problems may believe that they have adequate controls in place and, crucially, can trust their people. But an out-of-the-blue revelation of missing funds, a customer accusation of malfeasance, or a sudden government investigation can blindside the organization and leave leaders scrambling to halt a fraud scheme and contain the damage.
2. Red flags are often missed
   An organization may understand fraud drivers but overlook faint indicators or “traces” that hint something out of line could be happening. For example, advanced data analysis conducted by an insurance company investigating possible sales fraud identified potential employee red flags such as compliance violations, missed training, customer complaints, and expense report issues. Individually, the transgressions were viewed as minor offenses that didn’t warrant further investigation. However, combined in the profile of a particular employee, these indicators reliably pointed to instances of sales fraud in the ensuing months.
3. Effective fraud analytics rely on both technology and human dimensions
   Many organizations increasingly recognize the value of being able to predictively identify current activities that could grow into future problems. They are using analytics to help uncover and address these risks. Importantly, effective analytics involves more than summarizing and aggregating data. The value of the efforts hinges largely on whether the right indicators are being analyzed. A misdirected investigation can be of little use, yielding few insights and perhaps wasting limited resources in the pursuit of false positives.
4. Remember that fraud analytics is science and art, not science fiction
   Data analytics are revolutionizing fraud investigation. Many organizations hesitate to make the leap, however, because they may believe such capabilities are not yet mature or they fear doing so could lead to privacy issues. For example, false positives can still be a very real concern, especially when the latest generation of analytics tools give investigators the ability to analyze 100 percent of data sets rather than sample just a small portion, as in the past. Having the skills and experience to reduce false positives yet still derive meaningful insights from the data is an important requirement.
5. Two recent examples further illustrate the point
   The experiences of two companies in different industries demonstrate the value of investigating behavior patterns to uncover fraud:
   1. Major US utility company. A recent fraud event involving senior employees prompted the company to find ways to strengthen its fraud-detection analytics capabilities. With assistance from Deloitte, the company developed an analytics pilot within its customer operations center to test the capabilities of forensic analytics.
   2. Major investment manager. Increasing regulatory pressure and a desire to develop more proactive testing prompted this firm’s chief compliance officer to gather advice and recommendations on the development of an advanced analytics program. A dozen interviews and workshops with the firm’s compliance, technology, and internal audit departments produced valuable results.

Back to top

Our take

Organizations operate using familiar and comfortable patterns of behavior. However, rapid advances in technology and processes now offer extraordinary new avenues to enhance the sensing, analysis, and ongoing monitoring of fraud threats.

The involvement of specific and focused people within the organization, including professionals from the lines of business and the compliance organization, is critical to capitalizing on these developments and uncovering unknown and emerging risks. It is also important to not become disheartened by a few false positives or to neglect to follow up on leads.

Back to top