Capital one

docx

School

Arizona State University *

*We aren’t endorsed by this school

Course

502

Subject

Information Systems

Date

Oct 30, 2023

Type

docx

Pages

Uploaded by nickitheestallion

Capital One One way the credit card industry adds value is that they are filled with rich information about a customer (externally) since every interaction is recorded and captured. This allows the credit card company to have much information on their consumer and prospective consumers. Another way the credit card industry adds value is that it lends itself to being involved in massive scientific testing. Since the credit card is an increasingly flexible product and the industry has millions of customers, they realized that they could customize the credit card and its terms to fit each individual’s needs and provide them with unique offers if needed through direct marketing. It is also important to note that the credit card industry can be very profitable if individuals pay the company back. Members of the credit card industry can build this profitability if they manage to de-average the underlying steep gradient of consumer profitability. Capital one’s strategy was to deliver “the right product to the right customer at the right price.” This strategy goes further than FICO can measure because Capital One uses more data (through these tests and developed statistical models) that FICO does not have access to. Since it is the industry standard to use FICO, they saw that not all credit offers were representative of all people with the same FICO score when Capital One incorporated it. As a result, Capital One was able to utilize its “big data” capabilities to make use of the predictive power of specific variables to make risk predictions that significantly outperformed FICO. Combining this model with a response model allowed Capital One to market products to that optimal mix of creditworthy and credit- hungry consumers. Capital One ran tests on message strategies, prices, credit lines, product features, and acquisition channels while holding FICO scores constant. This helped Capital One realize that although individuals may have the same FICO score, these respondents can have various levels of risk associated with different offers. These tests allowed the analyst to learn the best marketing approach and

strategies for dealing with these situations (i.e., if a credit offer from a second channel results in a charge-off twice as high as the first channel). As a result, they could take this information and make the necessary adjustments to offers before distributing them nationwide. Capital One integrated “big data” by operating numerous tests (over 13,000 tests in 1997 alone and by 2000 in the tens of thousands) to gather information on their existing customers and their responses to these offers. Once they had this data, the company created a proprietary database. It was able to build statistical models so that they could instantaneously make actuarial calculations regarding the consumer’s lifetime NPV and predict consumer behavior. Leveraging these models also helped show how the company can minimize adverse selection. Using the statistical design, every test was set up to produce a meaningful, significant statistical result to help improve strategies while managing costs. This testing also allowed the company to realize when it might be time to exit a market. Through its tracking capabilities, they could monitor when products were no longer profitable (such was the case with balance transfers). Capital One developed a company of constantly evolving ideas. By introducing the complexity of mass customization and direct marketing into their strategy, they provided themselves with a better opportunity to make profound organizational changes. Capital One utilized data, both internally and externally, to assess the risk profile of potential consumers. These external credit bureaus provided these credit issuers with standard information. Through an algorithm developed by FICO, they could summarize each customer’s credit history and determine the level of “creditworthiness,” which they refer to as the FICO score. This might not be as effective as they’d hoped because the FICO scores addressed two main issues. First, this information is shared and readily available to all their competitors. The second is that consumers responding to the credit offers from the company weren’t completely reflective of everyone with the same FICO score. They found that compared to the non-respondents, the respondents were systematically more likely to be credit hungry and pose a greater risk than FICO predictions suggest.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version