Simple Score is an expert scorecard, based on common practices used in underwriting. Simple score is designed to measure the credit risk level and it outputs a single value that can be used as a proxy for how likely a customer is going to settle their liabilities.
The simple score can be used as a stand-alone score that is useful for companies that do not have enough resources or data to develop custom credit-scoring models, or as a machine learning feature in a scoring model.
Simple score is an integer in the range
10 represents the lowest risk level and vice versa.
The score indirectly reflects likelihood of customer settling liabilities, for instance, repaying loan, paying for bills or rent, and similar. It is expected that most of the customers who got the score 0 would not settle their liabilities, while the most who got score
10 would do so.
Knowing this, it is important to highlight that the response values may vary from country to country and from product to product, meaning what a
7 means to one use case and one product might not be the same for another use case, which is why there are no standard ranges for interpretation.
Example of Simple score distributions depending on the quality of clientele is illustrated below:
Lender 1: default rate 10% Lender 2: default rate 40%
As long as the bank statement has at least one transaction, Simple score will be calculated, however it is recommended that the statement holds at least 3 full calendar months of recent data. That would hold enough information of any outgoing loan payments, regularity in salary and other key risk indicators.
There is no upper limit for count of calendar months that could be considered too much data for Simple score calculations.
Simple Score is built from features that have aggregated various financial behaviour information. Behaviours can be explained and have importance as follows - income (20%), loans (20%), account activity (5%), expenses (15%) and other patterns (40%).
Both positive and negative angles from a customer's bank account data is used, for example, stable income and ability to pay utility payments regularly as a positive trend, while frequent gambling activities as negative aspect.