Fraud investigations go beyond the images of private detectives sleuthing surveillance and suspect interviews. The majority of cases are solved through the interpretation of documents and business portfolios. Data analytics plays a critical role in the work of a fraud investigator.

Data analytics is the science of examining raw data with the objective of drawing conclusions about that information. Data analytics is used across many industries to enable corporations and business entities to make better business decisions. In academia, particularly the sciences, data analytics are conducted to verify or disprove existing models or theories.

Screen Shot 2013-11-13 at 16.09.41It is noteworthy to mention the key differences between data analytics and data mining. Data analytics is distinguished from data mining by the purview, purpose and focal point of the analysis. Data miners sort through enormous data sets using sophisticated software to identify undiscovered patterns and establish hidden relationships. Data analytics, in comparison, focuses on inference, the process of deriving a conclusion based solely on what is already known by the analyst. Banks and credit cards companies, for instance, hire investigators to analyze withdrawal and spending patterns to thwart fraud or identity theft. E-commerce companies examine online traffic or navigation patterns to gauge which customers are more or less likely to buy a product or service based upon prior purchases or viewing trends. Modern data analytics often use information dashboards supported by real-time data streams.

Conventional methods of data analysis have long been utilized to uncover fraudulent activity. Investigators for the Securities Exchange Commission (SEC), for instance, use data analytics to identify potentially fraudulent activity such as front running, insider trading, fraudulent investment performance reporting, and window dressing. They are essential elements to complex and time-consuming investigations that deal with different domains of knowledge; these can include financial, economic, business practices and legal aspects. Fraud often consists of many incidents involving repeated transgressions using the same method. Fraud instances can be similar in content and appearance but usually are not identical.

Two forms of data analytics are frequently utilized in fraud investigations; these are link analysis and aberrational performance detection. Link analysis is a process which looks for relationships between two disparate data sources; this is particularly important in investigations involving insider trading. Using link analysis software on thousands of lines of data, investigators are able to quickly detect all instances in which the two suspects had a phone call with someone in common. This is a great improvement on the older non-technical methods, where investigators would have to attempt to look at each record individually.

Aberrational performance detection is a form of data analytics which focuses on unexpected performance to both identify fraudulent activity candidates as well as to ensure compliance with certain regulations as part of an examination. An example of aberrational performance detection can be exemplified in an investigation into a suspicious hedge fund. In this hypothetical scenario, the hedge fund’s recorded results were significantly better than its peers throughout both good and bad markets. Investigators, through such detection, would be able to determine that the hedge fund’s actual performance was significantly worse than its peers. Aberrational performance reviews such as these have been critical in identifying intentional valuation misstatements, Ponzi schemes, and other alleged illicit activities before they otherwise would have been discovered.

Screen Shot 2013-12-04 at 16.51.57To exemplify the use of data analytics in fraud investigations is the case of a Subsidiary Company and HO. In this case, an investigation was being conducted into the allegations of irregularities of inventory loss, vendor payments, and employees’ expense reimbursements. In a situation such as this the first step would be Data Acquisition to obtain digital evidence, such as the fuzzy invoices and duplicate expense reimbursements, making sure to follow safe acquisition methods to protect the validity of the evidence and investigation as a whole. This would provide you with some strong analytical results to help join the dots and create some substantial conclusions, such as the funding that there is a correlation between the shift manager and the inventory leakage. From this, one can now understand, interpret and optimize the results to derive relations and profile vendor fraud risk to identify the altered payee; answering the underlying question of who the payment has gone to, and thus solving the case.

Fraudulent activity can take many shapes and sizes. However, data analytics is an investigative tool that can provide results in almost any examination. There is, after-all, no such thing as the perfect crime. Through methods such as data analytics, fraud investigators are able to seize the critical evidence that is inevitably left behind. Fraud investigation and data analytics certainly go hand in hand.