Creditstar

My Role

As the lead designer in a team of three, I led the project throughout its entire lifecycle. My responsibilities included guiding the research phase, defining the user experience strategy, overseeing UI design, and managing the final handoff to the client's development team.

The Challenge

How might we create a system that predicts if a customer will default on a credit payment, especially when people often take up credits they cannot pay back?

Research & Discovery

According to Investopedia, a default occurs when a borrower is unable to make, misses, avoids or stops payments. Through interviews with bankers, loan officers, and account officers in Nigeria, we validated the problem and uncovered key insights:

• Widespread Problem: 100% of the financial professionals we interviewed confirmed they regularly face issues with credit defaulters.
• Ineffective Solutions: 67% stated that current methods, like disposing of collateral or contacting guarantors, were not consistently effective.
• User Willingness: After validating the problem, 100% of potential users showed strong interest in a predictive solution and were willing to provide the necessary data to use it.

Key Pain Points

User Personas

The Solution

Considering the Nigerian financial system lacks a universal credit score but uses a unique Bank Verification Number (BVN) for each individual, we designed a solution to leverage this. We created a data-parsing system that analyzes customer information to make an informed prediction.

1. If a user has no credit history, a prediction is made based on the turnover from their bank statement (minimum of 6 months).
2. If the user has a credit history, their repayment status is checked for any past defaults.
3. Demographic data such as age, marital status, education, and employment status are factored in to determine the credit limit for customers without a prior credit history.

Design Process

Information Architecture

Sketches

Usability Study & Key Findings

We tested the low-fidelity prototypes with experienced loan managers and individuals who had previously taken credit in Nigeria. The feedback was crucial and led to the inclusion of several key features:

• Send automated response messages to applicants.
• Ability for institutions to view a list of all applicants from their portal.
• Functionality to export applicant data for reporting.

High-Fidelity Prototypes

Style Guide & Components

The Impact

The platform was released in 2021 and is currently being used by two native banks in Nigeria, along with several other loan companies. The implementation of Creditstar has had a significant, measurable effect on business outcomes.