Stopping financial fraud before it happens is a better approach than having to address the issues after the fact. In this regard, some have implemented tools and techniques of Six Sigma to tackle accounting and financial fraud.

Financial fraud comes in different forms. They include tax evasion, money laundering, embezzlement and using money (especially in public agencies) for something other than the intended use. It also includes insurance fraud, credit card fraud and fraudulent internet transactions.

It’s a frequent issue for businesses, nonprofits and government agencies. But much like Six Sigma can apply in financial areas such as bank loan processing and increasing financial institution’s profit margins, it can also help detect fraud.

How to Detect Fraud Early

While automation has created a digital trail that makes fraud easier to detect, much of that detection happens after the crime has been committed. That makes it more difficult for victims of fraud to get their money back.

Six Sigma provides tools that can create a system in which standardized processes lead to earlier fraud detection. One of the most useful tools, according to the San Antonio Express News, is DMAIC, an acronym that stands for define, measure, analyze, improve and control.

It’s typically used to make existing processes more efficient. It can apply to fraud detection by following these guidelines.

  • Define – Define the sources of fraud, based on a company’s experience or known patterns of fraud from other cases
  • Measure – Measure the levels of fraud using verifiable data to ensure identification of all weak spots in a financial and accounting system
  • Analyze – Analyze data to determine how fraud happened, how long it took to detect, and whether the company recovered funds
  • Improve – Improve current measures to detect and prevent fraud, based on the data analysis. Choose interventions tailored to the company’s needs
  • Control – Monitor the new interventions to verify effectiveness and update as new fraud trends emerge

The Power of a Predictive System

study from the University of St. Cyril and Methodius in Slovakia also found the data mining used in Six Sigma beneficial to fraud detection. They place fraud detection in the control phase of DMAIC, after a revamped process is in place and performance is evaluated.

The researchers wrote that because of the predictive function of such Six Sigma-built systems, and the focus on reducing variance, any fraud would quickly get noticed because of a departure from expected outcomes.

In an example of putting Six Sigma to use in fraud detection, the Canadian Imperial Bank of Commerce used Lean Six Sigma to find waste and reduce cycle times in its debit card fraud prevention processes. The project ended up reducing the time between the initial alert of a fraud incident to completion of a case by 80%, as well as cut the average loss per incident by 45%. The program reduced debit card fraud losses by more than $7 million.

The company credited putting Lean processes into place that detected fraud faster and led to the bank acting on affected cards more efficiently.