“To gain market share despite growing competition, online lenders need to apply increasingly sophisticated rules and algorithms to improve portfolio metrics and the applicant experience.”
After an application for credit is submitted, lenders first assess the quality of the lead and apply low-cost knock-out criteria. At the first applicant data entry stage, consideration is usually given to minimize the information requested from the applicant. This is done to reduce friction in the initial application process, especially given the many online lending choices available at their fingertips. This is also done to address customer data privacy concerns. With as little applicant data as possible, lenders typically check if the borrower supplied information meets the minimum requirements such as age, loan amount and credit history. Common examples of requirements applied are: age less than 18, resident state of applicant, income, capacity, employment, requested loan amount below a certain threshold, bankruptcy record in the last ‘X’ number of years, previous customer, previous applicant and minimum credit score threshold, or hit. This is typically completed in the form of a soft pull of an applicant’s credit and is even advertised as such to encourage an applicant to apply and not have their credit impacted. Implementing these rules help lenders minimize the cost associated with processing applicants. For those that pass these initial rules, some lenders provide tentative feedback to the applicant. This is done to lower the abandon rate and to use as encouragement to request additional information along the application process. In addition, these rules also serve to perform an initial low-cost ID verification, where social security numbers, names, address, phone numbers, OFAC and IP information is cross-checked and evaluated. If a red flag is found, an applicant may be declined, asked to resubmit information or escalated for manual review.
Applications that pass the first stage will then undergo further evaluation and verification. To further verify credit eligibility, online lenders may use various data providers to gather more detailed borrower information as basis for segmentation, rule setting and scoring. Data may represent traditional and non-traditional commercially available scores, custom build scores or standard and custom attributes. The use of Champion/Challenger aids lenders to assess the effectiveness of their strategies across important metrics such as bad rate and loan conversion volume and cost.
The process will now also include more accurate, third-party data to verify the borrower’s identity, income, employment and bank activity. Part of this effort is to prevent first-party and third-party fraud. First-party fraud happens when a borrower applies for a loan with no intention of repayment, while third-party fraud arises when a borrower uses another person’s identifying information. Currently, lenders face another type of fraud risk which is loan stacking. This happens if the borrower has been approved in multiple applications without the lender being aware of the approvals and disbursement of funds. The use of KBA/out-of-wallet questions and loan stocking products are becoming popular with online lenders. An outcome of this stage is more hands-on manual verification and customer interaction. At the latter stage of the process, bank verification is done for funds transfer purposes.
Manual intervention to verify borrower data may yield more accurate results and may be necessary; however, this activity is balanced with the objective to lower the cost-per-conversion, as well as lower the overall time-to-fund of the loan. The latter is an important consideration because from the borrower’s perspective, time to access funds is an important customer experience component. From the lenders perspective, many moving parts go into time-to-fund, including, automated application decisioning, manual verifications, compliance and fraud risk. Lastly, rules to set maximum loan amount and pricing vary considerably between online lenders; consider factors such as risk, income and state, as well as a process to accommodate real time applicant feedback from offers and counter offers.
To compete more effectively, online lenders need to increasingly balance efficiency, accuracy and profitability. GDS Link’s technology and consulting has been helping online lenders do just that for many years.
About the Author: Miguel Florez, VP of Credit Risk and Analytics Consulting, leads GDS Link’s Risk Management Consulting and Analytics Group. The group integrates best practices, data and analytics to help companies meet their business objectives.[:]