Business Credit Rating Data

What Is Business Credit Rating Data?

Business credit rating data—sometimes called corporate or company credit rating data—provides an assessment of a company’s financial solvency and creditworthiness. It is a business’s credit score derived from several factors such as current outstanding debts, payment history, credit history, and loan payment delinquency.

Where Does Business Credit Rating Data Come From?

Business credit ratings are scored on the company’s payment history, credit history, size, current outstanding debts, and other relevant information. Essentially, the higher the score, the better the credit rating.

A small number of companies issue credit rankings for most companies in the world: Experian, Equifax, Dun & Bradstreet, and FICO. Equifax issues the Business Credit Report and Experian the Business Credit Score. Dun & Bradstreet’s most commonly used credit ranking is their Credit Score for Businesses, but their PAYDEX Credit Score is also popular. The PAYDEX score is calculated over the past year on the company’s payments to suppliers and vendors.

The FICO LiquidCredit Small Business Scoring Service takes the information issued by the first three companies and combines it to generate an average score. The Small Business Administration makes a decision on whether to approve a small business for a loan by using this FICO score.

Of course, there are other scores available, but these are the main ones in use.

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

Business credit rating models use a mix of structured and unstructured data. The structured data include firmographics, payment history, trade history, and public records like bankruptcies and liens. The unstructured data include news reports, company reports, meeting transcripts, and social media posts.

Developers are making advances in machine learning constantly, allowing the programs to process this unstructured data more quickly and efficiently. In particular, companies use natural language processing (NLP) to glean insights from these types of online sources. Social media users’ evaluations of companies, although highly informative are very difficult for computer programs to evaluate.

What Is This Data Used For?

Credit rating data determines the financial future of a company, from investments to loans to business credit cards.

AI models enable investors, lenders, and others to evaluate a company’s creditworthiness quickly. Then, provided the data and the model are high quality, investors and lenders can move forward with business.

Companies with less than desirable credit scores may also use the data to make a plan to raise their creditworthiness and become more valuable to investors and suppliers.

How Should I Test the Quality of Business Credit Rating Data?

Company credit data models that have been thoroughly tested and back-tested against last year’s pre-scored database may run well for years. However, a company’s situation or even the market or industry may change. Therefore, every so often, the model must be re-calibrated.

Essentially, the most important measure for an effective business credit rating model is how well its output stands up against the S&P Global Ratings Standalone Credit Profile. However, there are many additional tests for data quality that depends on your business goals for this data.

In addition to being complete and frequently updated, every good business credit rating model should have the business strategy defined at the outset. For example, if you are a lender and want to predict whether a company will default on a loan, you will need to carefully note the default status of each record and have access to both tradeline and non-tradeline accounts for the company. As another example, if you want to provide a loan to a new company that has little transaction or loan repayment history, the owner’s personal credit history is a good addition to your model.

Interesting Case Studies and Blogs to Look Into

Trust Science: Alternative Credit Scoring Resources
Credit Analytics Statistical Models’ Backtesting and Recalibration: A Primer

Tangible Examples of Impact

ZestFinance, in cooperation with Microsoft Cloud, unveils the ZAML Suite to enable lenders to more easily use the more effective machine learning programs to underwrite credit services, “potentially cutting billions in losses for banks and broadening access to credit for tens of millions of people.”
ZestFinance To Deliver First Fully Explainable Artificial Intelligence Solution For Credit Underwriting with Microsoft Cloud