Make more efficent and easy credit decisons

The new improved score is accompanied with an individual probability of Default (PoD). This shows the risk of doing business with a company in the form of percentage likeliness of insolvency. Commentaries will also be added to help explain the underlying reason for the PoD and score.

Creditsafe’s scorecard have an assessment scale of 1-100, divided into five risk bands, where each increased score implies a reduced risk. The scorecard’s high predictability means that companies with a low risk of insolvency are categorized with higher scores and in turn that companies with a high risk receive a lower score.

Each company is assigned a score, which is determined by an underlying risk forecast calculation. The output of this calculation is a percentage which we refer to as the Risk of Insolvency below. Each score from 1 to 100 corresponds to a range of “Risk of Insolvency” percentages as outlined in the below table.

In  order to improve the understanding of how a company's score is calculated, we have added commentaries to the score. The commentaries highlight the most important things for the current score and provide an insight into how the company's predict risk has been calculated.

The colour indicate on whether it is positive, neutral or negative. The comments are made by Creditsafe and are significant for understanding how the risk associated with the business is determined.

Examples of commentaries:

• This company has no auditor remarks
• This company has a lower net profit compared to similar companies
• This company has not changed its address or phone number in the last year

Companies come in many different shapes and sizes. Understandably, these businesses will present a different set of characteristics when they are heading into difficulty and have a strong likelihood of failure. The segmentation aims to define a set of sub-populations that, when modelled individually and combined, rank risk more effectively than a single model on the overall population in each country.

The main base for segmentation of the Danish population is the size (in terms of the total assets level) and financial accounts. The assessment is then based on the factors that have been proven the most statistically indicative of failure within each segment.

For start-ups and smaller companies, directors and ownership information play an important role, while information from the annual accounts in larger companies is more important for the risk assessment.

Companies with financial accounts:

• Micro enterprises
• Small businesses
• Medium-sized companies
• Large companies
  

• Newly started companies (that have not submitted their first annual accounts)
• Other companies where annual accounts are not available (e.g. sole proprietorships)

• Age of the company
• Industry
• Annual accounts
• Director's current and previous involvement in other companies
  
• Changes in the form of address and telephone
• Changes in the board of directors
  
• Auditor's comments
• Ownership

Our scorecard is complemented by a number of policy rules where we do not give a score between 1-100, but instead a comment. These rules apply, for example, when a company ceases or is exposed to high or difficult-to-assess risk or when we cannot assess the company with our generic scoring models. If the company does not have a score, an explanation is always given as to why the company cannot be risk assessed.

Examples of policy rules:

• Under bankruptcy
• Under voluntary liquidation
• Under reconstruction
• Insolvency information
• Account too old/not delivered
• No score for this business type
  

Gini is a statistical measure used to evaluate how powerful the Creditsafe scorecard is in predicting which companies will fail within the next 12 months. The Coefficient ranges from 0.00 to 1.

A 100% Gini is only in theory, but the closer to 100% the better. Gini for the danish model is 81%. The graph below shows the performance of the new scoring model and the red line at the beginning of the graph shows how we catch the bad companies on the lower part of the scoring scale. The grey comparison bar shows no separation between the good and the bad at each level.