Financial Fraud’s Foe is Data Analytics. So Why Are We Still Sampling the Haystacks?
Let’s say you are in the market for a new vehicle. How are you going to go about your research?
Here’s one way: Go down to the local laundromat and find a copy of Autotrader for listings of used cars up for sale. Or perhaps you’ll head to the newsstand and grab a physical copy of Consumer Reports for expert reviews on the latest models. Maybe pick up the phone and call friends and trusted peers for their thoughts. Head into a dealership to chat with a salesperson. Or even take an ad hoc survey of the roads around you and qualitatively deduce which models are popular.
But that’s not what you’re going to do, is it?
These are all valid research activities that can support your decision on which vehicle to buy, but let’s be honest: this day in age, that’s not how you are going to start your journey. You are going to use the Internet, like the rest of humankind.
In a few keystrokes, you’ll have access to the best prices and the most in-depth reviews of cars on the market today. Through social platforms you can effortlessly connect with local new and used vehicle groups. Dedicated auto sales sites will let you input your specific parameters (e.g. cost, size, colour, etc.) to quickly and easily determine which vehicle is right for you. There is definitely a place for the research activities laid out above, but in the context of the information-driven, technology-centric world we live in, it would seem rather archaic to start your buying journey and, indeed, attempt to make an informed purchasing decision based on such manual, time-consuming activities, right?
Sampling the Haystacks for Needles?
The same logic can be applied to the world of fraud. Sample-based approaches to fraud detection, analysis, and prevention have long been the go-to script for auditors and fraud examiners. And for a long time, that made sense. Large corporations that deal with the biggest cases of financial fraud simply have too much data to manually review and process. As a result we have conventionally turned to sample data to detect fraud. But if detecting fraud is like finding a needle in a haystack, a sample-based approach can be likened to looking at one haystack in a field of dozens, or perhaps it is more like taking one bale of hay from each haystack and drawing conclusions about the whereabouts or existence of the sought-after needle.
That sounds like an absurd approach, but it’s a fair analog for how financial fraud is approached today. Of course, we don’t have sweeping, automated tools for finding needles amongst haystacks, but in the world of fraud, we have such tools at our disposal in the form of data analytics.
There’s a Better Way
In a nutshell, data analytics helps us make sense of massive amounts of data by letting machines do what humans simply can’t: analyze every piece of data that could provide an indication of fraud. Data analytics represents a game-changing approach in the world of fraud detection, and even though it has been around for around 30 years (at least in our case), the uptake of this arguably necessary tool is remarkably low. For some reason, the internal audit function in general has been very slow to adopt new technology in general, and similarly either reluctant to embrace data analytics in the fight against fraud, or unaware that such tools are at their disposal.
So where to begin? I’m no fraud examiner, but I do have some of the best minds in auditing and fraud detection all around me here at IDEA, where data analytics for fraud investigations has been central to our offering for over a quarter century. That’s why I put together this very basic introduction to the world of data analytics and fraud detection. If you know data analytics will better equip you to detect and fight fraud in your or your client’s organizations but aren’t exactly sure how or where to start, this is a great place to start (see the download link below).
As with that new car purchase, it doesn’t make any sense to attack fraud with a 20th-century toolkit in a 21st-century world. The great thing is, you don’t have to, and this guide is your first step to getting a better handle on fraud detection and finding the many needles that surely lay hidden amongst the field of haystacks before you.
Paul Leavoy is a writer who has covered enterprise management technology for over a decade. Currently, he researches and writes on data analytics and internal audit technology for Caseware IDEA. Contact Paul directly or follow @CasewareIDEA to learn more.