In the financial markets of Korea and abroad, big data is being utilized industry-wide, starting from marketing and risk management to credit scoring. Credit card companies are utilizing big data when developing marketing strategies and new products by not only analyzing recent purchase history, but also social media and location-based information. Banks and insurance companies are turning to big data for risk and security management. In this chapter, we will take a look into big data use cases in finance as well as some lesser known cases from the fintech sector.
Big Data Use Cases of Credit Card Companies
As the credit card market almost reaches saturation, many global and domestic credit card companies are utilizing big data in marketing to create new sources of profit. Credit card companies are using CLO (Card Linked Offer) as a powerful marketing tool by combining customers’ needs identified using big data analytics with information collected their smartphones.
Multinational financial services corporation Visa offers RTM(Real Time Messaging) services that deliver location-based discount coupons by analyzing customers’ purchase patterns and tendencies by keeping track of when and where a customer purchased a particular item in real-time. As a result, Visa was not only able to increase customers’ credit card usage, but also secure new customers for their retailers. Visa has also developed and deployed a credit card fraud prevention system that analyzes customers’ credit card usage patterns with big data analytics.
Amex has received positive customer feedback for their personalized, social media based marketing campaign. With Amex Sync, customers can connect their social media accounts to their AMEX credit card to receive personalized discount offers. For example, if a customer likes a brand or restaurant on Facebook or Twitter, companies are able to provide discount coupons and relevant information directly to the customer, creating more targeted marketing opportunities than ever.
Korea’s leading credit card companies, Samsung Card and Shinhan Card, are also employing big data services.
Samsung Card offers LINK services, a personalized discount offering service based on big data analytics.
With LINK, customers are presented with discount offers from retailers they are most likely to visit based on their previous credit card transactions. Moreover, the discount is automatically applied during payment without customers having to separately present the coupon.
Compared to standard text message or target marketing, LINK lead to higher rates of actual purchase while also bringing noticeably a higher number of new customers to retailers.
Shinhan Card also successfully utilized big data when developing their latest offering.
Shinhan Card released a “Code 9” credit card series that clusters male and female clients into 9 groups based on credit card usage, consumption patterns, and preferred trends.
The company announced that the number of issued Code 9 credit cards recently surpassed the five million mark. Code 9 also boasts an average 10 % higher use rate compared to other major Shinhan Card credit cards. Code 9 is widely recognized as a successful case of utilizing big data in the industry.
Big Data Use Cases in Banks and Insurance Companies
Banks and insurance companies are using big data for risk management and managed security services.
Companies are applying big data to their operations when developing new products in order to prevent any potential loss that could arise from corruption.
With big data analytics, insurance companies are also able to release new UBI (User Based Insurance) products that are not only based on users’ basic personal information, but also user activity, market environment, and market trend analysis.
JP Morgan Chasedeploys big data analytics to detect fraud by not only analyzing their staff’s internet search history, but also their personal information including e-mails and call history. JP Morgan also utilized big data for determining an optimized real estate price determination model to be used when selling the property they acquire as collateral.
Big data can also be used to minimize social loss by analyzing the real estate market of reach region to determine the most marketable price of a property so that it can quickly be sold before a debtor becomes insolvent.
Auto insurance company Progressive leverages big data by amassing a trove of driver data with its driving tracker device in order to analyze driving habits and predict future accidents.
Using driving tracker devices, Progressive offers a usage-based, “pay as you go” insurance. That means the safer you drive, the bigger discount you get. This also promotes safe driving habits to customers.
Compared to other sectors, there are not many big data use cases from Korean banks or insurance companies. Most big data use cases are for cyber security and insurance fraud detection.
KEB Hana Bank leveraged big data to strengthen their cyber security. By developing a system that collects and analyzes log data using big data, the bank is able to prevent malicious code attacks and intrusions.
Samsung Fire & Marine Insurance successfully leveraged big data for insurance fraud detection. IFDS (Insurance Fraud Detection System) automatically detects high risk frauds, notifies on-site agents, and drives the investigation process.
For example, let’s say a customer files a stolen vehicle claim for their luxury car. The IFDS system automatically collects and analyzes customer and accident information and classifies it as a high risk fraud. Based on this classification, an agent conducts an investigation, discovering that the customer had secured a loan with the car as collateral and had filed a fraudulent claim in order to avoid repayment.
Big Data Use Cases by Fintech Companies
According to Wikipedia, fintech, which is the combination of the words “finance” and “technology”, is the new technology and innovation that aims to compete with traditional financial methods in the delivery financial services.
With the emergence of mobile banking and mobile wallet, fintech companies are offering differentiated services from existing financial companies. Major fintech companies such as PayPal, Lenddo and Ondeck don’t have access to financial transaction data that standard financial companies are in possession of. So instead, they utilize big data analytics for non-financial information by collaborating with social networking services or e-commerce companies. Fintech companies are not only analyzing customers’ shopping and consumption patterns, but also attempting to analyze their personality using psychology.
Lenddo, a Singapore-based software-as-a-service company, has developed a credit scoring algorithm that uses not-traditional online data in order to ascertain customers’ financial stability. With Lenddo’s algorithm, your credit score falls if you are online friends with a delinquent or frequently use negative words such as “car accident” or “unemployed” on your social media. In the case of small businesses, Lenddo ascertains credit rating based on the business’s reputation and activities.
German online lender Kreditech offers loans based on non-traditional data sources such as Facebook, eBay, and Amazon. Research has shown that a person who makes spelling errors on their loan application is more likely to be overdue on their payments, so Kreditech uses the number of spelling mistakes on an application when determining creditworthiness. In addition, how thoroughly a person reads the general terms and conditions for loans is another factor they consider. That is because the more thoroughly a person checks the terms and conditions, the more likely they are to repay their loan. Another interesting factor the company utilizes for credit scoring is how often a person receives package deliveries. The reasoning behind it is that if a person regularly online shops, they are most likely to have a steady source of income.
In Korea, it is not easy to find big data use cases in the fintech industry. One of the few use cases comes from peer-to-peer lending service provider Lendit. Lendit uses a big data credit score model that considers borrowers’ Facebook information as well as their action patterns when reading the investment instruction on Lendit’s website. For example, a person who recently changed jobs to a venture company with lower pay applies for a loan. Analyzing the applicant’s social media reveals that the applicant and his family are professionals with a stable source of income. In this case, big data analytics reveals that the applicant’s ability to pay score is higher than his income score, allowing him to receive a loan on better conditions.
Major Korean financial companies mainly use big data analytics for structured data. However, by including social media data and unstructured data in their analytics, business can create new products and business models that can stay ahead of changing customer needs and market conditions. If banks and credit card, insurance, and stock companies begin utilizing big data analytics, which has already been proven in the fintech industry, they will be able to see big results in terms of product development and risk and security management.
Data collection and usage is much more restricted in Korea compared to other countries. Pursuant to the Personal Information Protection Act, personal information cannot be utilized in big data analytics. In order to overcome these limitations, the Korean government released a set of guidelines that state de-identified data does not fall within the legal scope of personal information. These guidelines are expected to positively impact the use of big data by various industries including fintech, as well as cultivate new big-data based business opportunities.
In the next chapter, we will take a look at use cases from the biotechnology industry.
※ This is an article from the 2016 spring edition of KIIE’s (Korean Institute of Industrial Engineer) IE magazine, and is republished with the KIIE’s approval.
▶   The contents are protected by copyrights laws and the copyrights are owned by the creator and Samsung SDS.
▶   Re-use or reproduction as well as commercial use of the contents without prior consent is strictly prohibited.
After receiving her Ph.D for industrial engineering from Gorgeia Tech in 2009, Dr. Seoyeon Kim worked as an industrial engineering researcher at the National University of Singapore before joining Samsung SDS in September 2010. As a data scientist, she leads multiple big data projects while also working as an instructor for an in-house data scientist training program she helped create. Dr. Kim regularly shares her in-depth big data expertise as a contributor for CommonSDS and IE magazine and also actively participates in various industry seminars.