Data Analysis

CRF data analysis, in a broad sense, is referring to data integration, data mining including risk and sales scorecard development, customer segmentation, business strategy development and implementation, as well as database marketing development.

Data integration. Most of the data held by companies, banks and financial institutions is in the form of raw data, or operational data. It becomes useful information for decision support only after a huge amount of data integration work and intermediate attributes development work. CRF utilizes international consumer finance knowledge and experience, combined with local operations and a local execution team, to develop knowledge-intensive data products for the Chinese consumer industry including consumer finance.

Data mining. Through tera-data integration, modern statistics and non-statistics methodology, the business analyst community works on data and gets the information for decision support. One technique is the score cards method. Differing from commercial loans, consumer loans, especially credit card loans have to deal with the small credit loans, but big volume processing. The scorecard method is a way to handle such volumes efficiently and consistently in approving application and managing credit limit by setting up a scorecard cutoff point, therefore reducing the processing costs. Customer segmentation and strategy is also based on the data driven approach. Its core is based on target marketing, i.e. to segment customers into different groups based on their behavior and in accordance differentiate the targeting strategy. CRF has the most advanced data mining methodology and experience, both internationally and domestically. Its core team members come from data mining and decision management teams from some of the top global financial institutions such as Citigroup, HSBC North America and Bank of America as well as Sears Credit. CRF is also a major player in the development of customer segmentation, scorecards, business strategy and database marketing for the financial industry as well as other industries in Mainland China.

Database marketing. Database marketing is basically referring to two things: one is data mining for marketing purpose, the other is applying the learning outcome from data mining to marketing and sales to find more customers for products and cross selling. It includes tera-data analysis utilizing modern parametric and non-parametric methods. It takes a multi-dimensional view on multi-dimensional data such as consumption, purchasing and payment behavior as well as demographic data. It tries to discover the cross correlation, seasonality, frequency, recency and monetary amount between different purchasing activities. The knowledge can then be applied to a customer database to find new customers for merchants.