Credit origination is the initial process of obtaining an exposure such as a credit card, loan, or mortgage. This involves multiple steps that can vary depending on the financial institution. However, the basic process contains the steps of application assessment, underwriting, granting approval, funding, etc. This process includes manual and labour-intensive tasks as well as automated tasks.
This critical process plays an important role in obtaining a loan or credit card. Therefore, there are many risks and challenges involved.
Challenges in Credit Origination Process
The goal of financial institutions is to make a profit while helping the customers to reach their goals. However, the steps such as assessing an applicant is a complex and challenging process. This has a direct impact on the underwriting process, costs, and approval time. That can also result in unforeseen losses on both the financial institution and the applicant.
How Decision Intelligence Can Help?
Decision intelligence can help to tackle the issues with predictive analytics and prescriptive analytics. Using predictive analysis, the lenders can foresee the risks that can occur while the entire business can be efficiently modeled using workflows and business rule engines in prescriptive analytics.
The steps are:
- Predict the repayment delay using a machine learning algorithm according to the applicant data
- Determine the repayment delay risk score according to the predicted repayment delay
- Finally, determine the interest rate
For the prediction, we have used a dataset of credit card repayments. Using this dataset, we predict whether the user is going to delay the repayments.
Based on that, the repayment risk score is determined using a decision table.
Finally, the interest rate is given as the output depending on the repayment risk score and the years of membership.
The example shows how predictive analytics along with prescriptive analytics can be used to determine the risk and then interest rate according to the applicant’s personal data. It gives a more justified decision at the end.
|Stages||How Decision Intelligence can Help|
|1. Pre-qualification process||Using predictive analytics, the risk can be determined according to the identity, income, and expenses of the applicant. Also, a rule engine can be helpful to indicate eligibility.|
|2. Loan/ Credit Application||Decision Robotics can be used to automate the application filling process.|
|3. Application Processing||Decision Robotics or a business rule engine can be used to check whether the application is complete and accurate.|
|4. Underwriting process||Using a business rule engine, credit score, and risk score can be determined with higher accuracy and efficiency. Predictive analytics can also be used to develop standards.|
|5. Credit Decision||Integrating a workflow can efficiently approve or reject the applications. This will be helpful to know the status of the process increasing transparency.|
|6. Quality Check||Before giving the final decision, accuracy and the quality of the decisions can be determined using a business rule engine.|
|7. Funding||As a part of a workflow, the authorities can begin the funding process by disbursing the amount to the applicant.|
Credit origination is a complex process with multiple critical steps involving assessing personal and financial data, underwriting, and disbursing funds. That is why decision intelligence should be used. From the lenders’ side, this will improve the productivity and the accuracy of the credit origination process. And, from the customers’ perspective, a more customized user-specific service will be received in a timely manner. Usage of predictive analysis to make foresee the risks and trends combined with prescriptive analysis to follow the business rules including human interaction as necessary can indeed streamline the process enhancing the overall customer experience while helping the lenders to make more insightful decisions.
Last updated November 10th, 2020 at 03:26 pm, Published October 22nd, 2020 at 03:26 pm