Having a massive amount of data on creditworthiness aspects of your business relationships is nice. But having one creditworthiness score is even nicer. How do you get from huge mountains of data to one score? An explanation in 6 steps.
An extensive creditworthiness report tells you a lot. But you can't see at a glance whether a potential new business contact is financially sound enough to do business with. See the power of creditworthiness scores: one figure on which you can immediately base a decision.
Altares Dun & Bradstreet customizes these credit scores. Not only do we build specific models for each country - with scores that are therefore created differently. But we can also create a customized scoring model for your company, based on your risk appetite. On the basis of these scores, you can, for example, calculate the automate onboarding of new customers. Or monitor the creditworthiness of existing customers in real time. real-time monitoren.
Step 1: Collecting data
With the Dun & Bradstreet Data Cloud we have access to the world's largest commercial database, full of valuable information such as payment behavior, collection information, corporate structures, turnover figures and balance sheets.
We get our data on (creditworthiness aspects of) your business relationships in various ways. Worldwide, we have more than 30,000 data sources. One of the most important data sources is our own. Within our DunTrade program, we obtain daily payment experiences from numerous companies around the world.
Step 2: Collect, verify, and structure data through the DUNSRight process
Through the DUNSRight-proces , we guarantee the quality of all our data. We collect, verify and structure the data in five different steps. So that all information is current, complete, reliable and globally consistent.
Step 3: Distinguish active and inactive companies
Our database also contains data on companies that are no longer active. First of all, because we want to keep track of these companies, for example, because they may have a relaunch. In addition, our historical database contains a wealth of important information, for example in the area of bankruptcies. What happens in the months preceding a bankruptcy? Do we detect similar patterns in companies that are currently active? We use these kinds of insights to calibrate our models, thereby continually improving our scores.
Step 4: Our data goes through the scoring models
All data elements such as industry, age of the company, financial data and payment behavior are taken as input for the scoring models. Each element has its own rating.
Step 5: Raw score
Added together, all those ratings form a raw score, which is the basis for the Bankruptcy Score.
Step 6: Bankruptcy Score
The result of these 5 steps is the Bankruptcy Score. This is what is known as a percentile score: a number from 1 to 100 that indicates how likely bankruptcy is, related to other companies in the market. For example, if a company's score is 80, then 80 percent of companies are more likely to go bankrupt. So a score of 1 means a relatively very high probability of bankruptcy, a score of 100 a particularly low one. Thus, the Bankruptcy Score helps companies make informed decisions.
Our scores within your organization?
Would you like to use our scores within your company to avoid risks, identify opportunities and make better decisions? Check out our solutions for credit risk management.
Want to know more about the possibilities of automating credit decisions based on multiple data sources? Then check out our solutions for automating the customer acceptance process.
