DIGITALIZATION OF CREDIT RISK ASSESSMENT IN RETAIL LENDING


As part of its regular risk management, the Bank implemented and operates on the best IT technologies and the best risk management practices. In mortgage lending and car loans 82% – 89% of all credit decisions are made automatically as well as 96% of credit decisions in consumer loans.

The Bank strives for the best experience in the market, so that a credit decision for a client who has applied to the Bank is made in real time (more 90% requests less 60 seconds to make a decision). For this purpose, the Bank uses both a set of traditional customer data and big data analysis tools. The Bank collects a large amount of information from the client's questionnaire, analyzes the client’s own transactions within the Bank, which is an additional element of behavioral scoring. In addition, the Bank takes into account customer information from government agencies, as well as their credit history.

A high-tech loan conveyor and smart scoring models have maximized the share of robotized decisions in taking on risk.

Mortgages

Consumer loans

Car loans


The Bank actively employs customer Internet activity data, which yet another element of transaction scoring, increasing model predictability. In order to cast away the questionnaire and make the loan approval as convenient as possible for the customer without worsening the accuracy of the estimated losses, the Bank uses both the customer-generated and its own data, as well as the most efficient external data sources.

Risk assesment tools: In-House vs Outsource