What can I do with a master's degree in Management Analytics? Part two: Fraud Analytics

Print | PDF

Information Superhighway Robbery

It’s fair to say the digital revolution left very little of our society unchanged. Since the advent of microprocessors and fiber-optic cables, we’ve enjoyed the speed and convenience of moving many of our daily tasks online.

Criminal elements haven’t missed the opportunity to profit from the extremely variable quality of security used in online transactions. No longer relegated to the shadows of dark alleys, waiting for the next wallet to walk by, criminals have moved online along with almost everyone else.

The sophistication with which online fraudsters operate has created a real and widespread problem in Canada. According to CPA Canada’s 2020 Fraud Survey, one in three Canadians have fallen victim to some kind of financial fraud. In the pre-pandemic days of 2019, already 74 per cent of Canadians routinely conducted purchases and other financial transactions online, with 45,000 of them losing nearly $100 million to criminal fraud that year alone.

“The frequency in which you carry out online transactions increases your risk, especially because Canadians are not using the most secure methods and systems,” says Claudiu Popa, principal risk advisor at Informatica.

Analysis from just six months into the pandemic found that Canadians had plenty of reasons to increase their online transactions to the tune of 250 per cent.

This kind of increase has put unprecedented pressure on banks to identify and stop fraudulent transactions that affect not only consumers with weak passwords, but even the banks themselves.

Big data to the rescue

Data science and business analytics are employed to solve business problems every day. This practice extends naturally to financial institutions flooded with information. Using business-centric data science and analytics practices to decrease financial risk in e-commerce is common. Time series anomaly detection can be used to conduct real-time fraud detection with what is known as ‘Card-Not-Present’ transactions.

To increase their capacity to sort through oceans of data, companies are tapping into the talents of recent graduates in order to expand their analytics divisions, especially in the fraud prevention field.

Britina Minjung Kim (BBA ’17) joined Scotiabank in November of 2020 as a senior manager of fraud analytics. She credits her natural predisposition toward math and her two undergraduate degrees in the double degree program in mathematics and statistics from the University of Waterloo and business administration from Wilfrid Laurier University as part of the reason why her career has taken off so quickly.

Skilling up to a future-focused career

“Transitioning from high school to university, I wasn’t exactly sure what I wanted to do, but I did know that I really loved math – I just wasn’t sure there was a direct career path that came out of it,” says Kim.

“My advice to anyone in a similar position is to keep your options open. As I went through my co-op terms, I realized how powerful the combination of knowledge in math, statistics, and business can be. This led naturally to my career now.”

Prior to joining Scotiabank, Kim was a consultant at Deloitte where she had the opportunity to wear different hats while working on a variety of data-related projects across a number of industries.

“Deloitte was great because I got exposure to all kinds of career paths, especially the ones open to me because of my ability to see the connection between math and business,” says Kim.

During her career exploration at Deloitte, Kim joined a project that had everything she was looking for.

Her team was engaged by one of the Big 5 banks to look for employee-driven inconsistencies. “Basically, we were looking for bad actors within the bank itself,” explains Kim. “I really liked the project because it brought together my stats and finance background. It was especially satisfying to help stop bad actors and the real losses they cause for the bank.”

By the end of the project, Kim knew it was time to pick a professional direction. She noticed more fraud analytics jobs opening up as companies started investing even more resources into those business units.

“The opportunity for growth right now stems from companies realizing just how much they can do with data to make better business decisions, especially around fraud,” says Kim. “We aren’t just talking about artificial intelligence or machine learning, because in the banking industry there are so many little things you can do with data analytics to make big business decisions – that’s where the growth lies right now.”

As large, historic organizations, Canada’s big banks have developed their own in-house solutions to manage the flow of information across business divisions, meaning the data remains mostly fragmented. Finding the right information, cleaning it up for proper analysis, and using those insights to drive business decisions are where the focus is currently directed.

“I find this work very satisfying,” explains Kim. “Whether you’re a visual person, or a numbers person, this work helps you see a much bigger picture of what’s going on and to be able to make decisions in an unbiased way, which is really an exciting way to analyze past events to try to predict future ones.”

Using a combination of core data gleaned from account information, customer demographics, and even IP addresses and dark web lists, Kim assesses the risk level of any given transaction.

“Fraud prevention essentially comes down to establishing a baseline for normal behaviour and from there we ask questions to gain business intelligence, such as: ‘how different is this transaction from the others?’, ‘is this email linked to known fraudulent activity?’, and ‘how do we proceed with this information?’” explains Kim.

Kim’s business preparation comes in handy throughout the day but especially when answering that last question. Purely technical analysts will often fail to see the direct value of these decisions or how to work best with other business units to solve larger, more complex problems.

“It’s like connect-the-dots,” says Kim. “The more dots you have, the clearer the picture.”

The Lazaridis School thanks Britina Minjung Kim for her participation in this story. For more information about the Lazaridis Master of Science in Management Analytics program, visit our website.