With the widespread application of mobile Internet, cloud computing, Internet of Things and social networks, human society has entered a new "big data" information age. The future of bank credit is also inseparable from big data.

Many domestic banks have begun to try to drive business operations through big data. For example, China CITIC Bank’s Credit Card Center uses big data technology to achieve real-time marketing, China Everbright Bank has established a social network information database, and China Merchants Bank has used big data to achieve real-time marketing. Data development of small and micro loans. From the perspective of development trends, bank big data applications can be generally divided into four major aspects:

The first aspect: customer profiling applications.

Customer profiling applications are mainly divided into personal customer profiling and corporate customer profiling. Personal customer profiles include demographic characteristics, consumption power data, interest data, risk preferences, etc.; corporate customer profiles include the company's production, circulation, operations, finance, sales and customer data, and relevant upstream and downstream industry chain data. It is worth noting that the customer information held by banks is not comprehensive, and it is sometimes difficult to draw ideal results based on the data they own and may even draw wrong conclusions.

For example, if a credit card customer swipes the card 8 times a month on average, makes customer service calls 4 times a year on average, and has never made a complaint, according to traditional data analysis, this customer is a highly satisfied customer. Lower risk customers. But if you see the customer's Weibo, the real situation is: the salary card and credit card are not in the same bank, repayment is inconvenient, several customer service calls have not been answered, the customer has complained many times on Weibo, and the risk of losing this customer is higher. Therefore, banks should not only consider the data collected from the bank’s own business, but also consider integrating more external data to expand their understanding of customers. Including:

(1) Customer behavior data on social media (for example, China Everbright Bank has established a social network information database). By connecting the bank's internal data and external social data, a more complete customer puzzle can be obtained, allowing for more precise marketing and management;

(2) Customer transaction data on e-commerce websites, such as China Construction Bank combines its e-commerce platform and credit business. Alibaba Finance provides unsecured loans to Alibaba users, and users only need to rely on their past credit;

(3) Industries of corporate customers Upstream and downstream data on the chain. If the bank masters the upstream and downstream data of the enterprise's industrial chain, it can better understand the development of the enterprise's external environment and predict the future situation of the enterprise;

(4) Others are conducive to expanding the bank's customer service Interest and hobby data, such as Internet user behavior data from the DMP data platform currently emerging in the online advertising industry.

The second aspect: precision marketing

Based on customer portraits, banks can effectively carry out precision marketing, including:

(1) Real-time marketing. Real-time marketing is based on the real-time status of the customer, such as the customer's current location, the customer's latest purchase and other information to carry out targeted marketing (a customer uses a credit card to purchase maternity products, the probability of pregnancy can be estimated through modeling and recommendations to pregnant women business that people like); or consider life-changing events (changing jobs, changing marital status, moving to a new home, etc.) as marketing opportunities;

(2) Cross-marketing. That is, cross-recommendation of different businesses or products. For example, China Merchants Bank can effectively identify small and micro enterprise customers based on customer transaction record analysis, and then use remote banks to implement cross-selling;

(3) Personalized recommendations. Banks can provide personalized recommendations for services or banking products based on customer preferences, such as accurately positioning customer groups based on their age, asset size, financial management preferences, etc., analyzing their potential financial service needs, and then conducting targeted marketing. Promotion;

(4) Customer life cycle management. Customer life cycle management includes new customer acquisition, customer churn prevention and customer win back, etc. For example, China Merchants Bank built a customer churn early warning model and launched high-yield financial products to retain the top 20% of customers with a churn rate, which reduced the churn rates of gold card and golden sunflower card customers by 15 and 7 percentage points respectively.

The third aspect: risk management and control

Including means such as risk assessment of small and medium-sized enterprises and identification of fraudulent transactions.

(1) Small and medium-sized enterprise loan risk assessment.

Banks can use the company's production, circulation, sales, finance and other related information combined with big data mining methods to conduct loan risk analysis, quantify the company's credit limit, and carry out loans to small and medium-sized enterprises more effectively.

(2) Real-time fraudulent transaction identification and anti-money laundering analysis. Banks can use basic cardholder information, basic card information, transaction history, customer historical behavior patterns, ongoing behavior patterns (such as transfers), etc., combined with intelligent rules engines to conduct real-time anti-fraud analysis of transactions. For example, IBM's financial crime management solution helps banks use big data to effectively prevent and manage financial crimes. JPMorgan Chase Bank uses big data technology to track criminals who steal customer accounts or invade automated teller machine (ATM) systems.

The fourth aspect: Operation optimization.

(1) Market and channel analysis and optimization. Through big data, banks can monitor the quality of different marketing channels, especially online channel promotion, to adjust and optimize cooperation channels. At the same time, you can also analyze which channels are more suitable for promoting which types of banking products or services, so as to optimize channel promotion strategies.

(2) Product and service optimization: Banks can convert customer behavior into information flow, and analyze the customer's personality characteristics and risk preferences, gain a deeper understanding of customer habits, and intelligently analyze and predict customers. needs to carry out product innovation and service optimization. For example, Industrial Bank is currently conducting a preliminary analysis of big data, mining and comparing repayment data to identify high-quality customers, and provide differentiated financial products and services based on differences in customer repayment amounts.

(3) Public opinion analysis: Banks can use crawler technology to capture relevant information about banks and bank products and services from communities, forums and Weibo, and make positive or negative judgments through natural language processing technology. In particular, they should grasp negative information about banks and bank products and services in a timely manner, and discover and deal with problems in a timely manner; positive information can be summarized and continued to be strengthened. At the same time, banks can also capture the positive and negative information of banks in the same industry, and keep abreast of what their peers are doing well, so as to use it as a reference for optimizing their own business.

Banks are credit businesses, and the power of data is particularly critical and important. In the era of "big data", modern information technology represented by the Internet, especially the vigorous development of new communication methods such as portals, community forums, Weibo, and WeChat, and the widespread application of mobile payments, search engines, and cloud computing, have built a It has created a new virtual customer information system and will change the modern financial operation model.

The characteristics of big data such as massive quantification, diversification, rapid transmission and value will bring new challenges and opportunities to the market competition of commercial banks. In the data age, only the wise survive. The future of bank credit is to win the future from data and gain stability from risk control.