Tech vs. Fraud: Five internal investigations best practices

April 21, 2020

A blog post by William Pollard, partner, Deloitte Risk and Financial Advisory and Samantha Parish, principal, Deloitte Risk and Financial Advisory

At a time when practically every detail, major and minor, is being captured in company data, businesses now can take a new, technology-enabled approach to internal fraud investigations: One that's more consistent, defensible, and efficient. Here's a quick look at five insights from our new report on tech-enabled ways to combat fraud:

Siloed is bad, integrated is good.
Different types of investigations are often necessarily conducted by different departments in an organization with limited, if any, coordination or communication among them. The resulting lack of integration can cause potential fraud risks to be missed, improperly identified, and under analyzed. Cross-functional processes and technologies – like shared case management software and data sensing tools – can help internal investigators identify, communicate, and analyze issues through a broader lens. And this can ultimately lead to a better understanding of the true nature and scope of the fraud.

Why the time is right for a tech-enabled portfolio approach.
While the use of technology in investigations isn't new, the rapidly expanding and accelerating role of technology and data in personal and professional lives is. When virtually every action people take is being captured digitally, finding the fraud in ever-increasing data sets and sizes is an inherent challenge. It becomes increasingly important to apply a diverse portfolio of tech-enabled investigative techniques and approaches – from business queries to digital due diligence to data analytics.

Tech-enabled investigations can be predictive value drivers rather than cost centers.
The types of behaviors underlying fraudulent activities can be reverse-engineered by investigations, data, and business specialists to develop predictive models. If a company clearly understands what constitutes "bad behavior" among its employees, vendors, and other stakeholders, the results of tech-enabled investigations can be used to design algorithms that can predict, detect, and ultimately help prevent that behavior.

The same techniques can uncover performance improvement opportunities.
It's not unusual for multinational organizations to conduct tens or even hundreds of simultaneous investigations that, on the surface, have no connection to each other. But what if forensic, data, and business professionals in an organization could connect the dots between investigations? What if data sets could be harnessed to identify patterns or trends across the organization? With new tech-enabled investigation capabilities, organizations can develop a broad-based risk profile supported by a library of risk issues that enables the shift from reactive to proactive – and eventually to predictive fraud risk management.

Even more opportunities exist for tech-enabled investigations.
With tech-enabled investigation capabilities, an organization can bring people, processes, and technologies together to seek out and respond to rare events that can cripple the organization. Organizations with predictive capabilities could find problems before they become systemic. And if a rare event does go undetected, the learning's from the event can be applied to preventive measures for the future.

In the fight against internal fraud and misconduct, it's important to remember that data science on its own won't make you victorious. Effective investigations require more than just the right technology support, they need humans with a strong understanding of data science, business, and criminal behavior.

To learn more about how your organization can harness investigation capabilities for accelerated performance, check out Deloitte's Tech-Enabled Investigations Spark Experience.

And please take a moment to comment below or send me a message. I'd like to hear your insights and experiences on using technology-enabled approaches to combat fraud.

Back to top