Credit Data

What Is Credit Data?

Credit data includes information on the ability of a customer to repay a loan.

The customer in question may be an individual or a company but the principles of credit lending remain the same. Additionally, individuals receive credit scores while companies receive credit ratings. 

Credit data includes the credit types and amounts taken on by the customer as well as the number of payments made. Additional information includes the terms of the loan or restrictions imposed on the customer due to nonpayment. Other possible sources of data include characteristics of the customer’s accounts at their provider bank or union.

Whether referring to a rating or a score, credit numbers are never static, with one missed payment bringing down even the best score. Likewise, credit takes time to build up and lenders want a customer who can repay loans consistently. In other words, a customer with good but short credit history is not as desirable as a customer with the same score but a longer history.

Where Does Credit Data Come From?

You can collect credit data from all kinds of sources. Common sources include utility bill payments, telecom payments (cell phone, cable, or satellite bills), rental payments, and bank account information. You can also find this information from digital footprints and full-file public records as well as local registries, local publications, debt collection sources, and more. Some data vendors also have the ability to analyze data on millions of invoices paid daily from millions of tradelines worldwide.

Trustworthy data vendors, like those on our site, not only collect data but also run it through scoring algorithms to generate personalized scores and ratings.

What Types of Columns/Attributes Should I Expect When Working with This Data?

Common credit data attributes include payment history, credit utilization, credit history, types of credit, and new credit.

Additional attributes include alternative credit data like employment history and income, property ownership, and history of bankruptcy, liens, and judgments. Even the contact information in cell phone records can be used.

What Is This Data Used For?

Businesses use this data to guide various business decisions. Investors, of course, use credit ratings to determine the risk of a potential investment. Other businesses use it to decide whether to partner with or acquire another company.

Finally, borrower companies also pay attention to this data. Credit ratings and scores not only determine the interest rate at which they must repay a loan but also whether they will be approved for a loan at all.

How Should I Test the Quality of Credit Data?

A good credit data set should be updated in real time. Vendors must regularly check the dataset for inconsistencies, errors, and duplications. They should also validate the data using techniques from data issue tracking, certification, and statistic collection to workflow management and aggregate-based verification.

What Are the Most Important Factors I Should Vet when Selecting This Data?

A reliable credit rating vendor offers 24/7 monitoring with real-time notifications of credit risk changes delivered directly to you. They should also offer information on the performance of current and previous company directors and shareholders.

Some vendors also provide training and support in account management.

Finally, while some vendors use their own scoring system, they should nevertheless include a credit limit endorsed by credit insurers.

Interesting Case Studies and Blogs to Look Into

Deloitte: Credit scoring – case study in data analytics
ID Analytics: Alternative credit data case study: Top-ten card issuer uncovers credit invisibles
Credit Scoring Through Data Mining Approach: A Case Study of Mortgage Loan In Indonesia
Analytics India Mag: Top Credit Scoring Startups In India That Use AI

Tangible Examples of Impact

A study conducted in the Bank for International Settlements has shown that machine learning-based credit scoring models outperform traditional empirical models (using both traditional and non-traditional information) in predicting borrowers’ losses and defaults. The machine learning models better predict losses and defaults following a negative shock to the aggregate credit supply.

“The credit scoring industry has been disrupted by alternative methods of collecting data in recent years. Instead of looking at an individual’s credit payment history, alternative credit scoring providers use data points from mobile phone usage or psychometric tests to determine the likelihood of an individual repaying their loans…this helps lenders expand their pool of borrowers to those with insufficient credit history while keeping risks in check.”

—The Edge Markets: TheWall: Alternative credit scoring gaining relevance

“By leveraging smartphone and web behavioral metadata, we make it possible to underwrite the credit applications of individuals including thin-files, millennials, self-employed, and gig economy workers in real time, with high predictability, great accuracy, and full privacy protection. As a result, clients have seen 20% higher new-to-bank customer approvals, a 15% reduction in non-performing loans, and a 22% dip in delinquency.”

—Mobile Payments Today: Thick file customers vs thin file: AI helps level the playing field