Creditstar is a fintech startup in nigeria creating B2B tools for financial organizations to synergize product and services. This project was contracted to waldroid consult to conduct a market research and design the product.
I was the lead designer in a team of three designers, a UI designer and a UX Researcher. I led the project throughout the lifecycle from the research to the handing off to the client for the developers.
How might we create a system that predicts if a customer will default a credit payment because people take up credits they cannot pay back.
According to Investopedia, a default occurs when a borrower is unable to make, misses, avoids or stops payments.
From our research which involves Bankers, Loan officers and Account officers, we were able to deduce the following:
- 100% of the population accented to the fact that they have faced the problem of credit defaulters.
- The prevailing methods of solving this problem are disposing of collateral and summoning the guarantor, and 67% indicated that these methods have not been totally effective.
After validating the problem through user research, 100% of the users indicated interest in a product that would solve their problem in this way, and the willingness to provide the necessary information to use such a product.
Existing unforseen red flags in clients’ application and financial information.
Inability to precisely predict customers’ liquidity and disposition of collateral.
Many clients default in their payment thereby exposing the bank to losses therefore, customer get credits above what they can pay back
After intensive and extensive research and considering that the nigerian financial system does not have a credit system and run only a unique number Bank Verification Number (BVN) per individual, this will help to check any credit history with other financial institutions. We created a form that parses the customer data to consider the followings:
- If the User have no credit history then some prediction will be made base on their most used banks with a statement of 6 months minimum to run a turnover.
- if the user have a credit history the prepayment status will be checked if there is a default history.
- Data such as age, marital status, education and employment status will have effect on the credit limit for the customer without credit history.
After the low-fidelity prototypes, the prototypes were sent to some loan managers and banker who are experienced in managing loans in their company. We also sent the forms to some people who have had experience in taking some credit in nigeria.
Some of these features were requested
- Send message of response to the applicant
- Check lists of applicants who submitted through the institution portal
- Export applicants data