GigPay, as a payment platform, has been experiencing increasing fraudulent transactions, resulting in financial losses and failed transactions. The objective of this analysis was to understand the overall business and fraud health, identify where fraud is concentrated, and determine the key areas that require attention to reduce future losses. I have approached the analysis through three stages: Monitor → Diagnose → Prioritize. First, I established the overall fraud health using KPIs and trends such as transaction value, fraud rate, fraud exposure, fraud loss, and affected customers. Then analysed fraud across country, channel, KYC tier, gig segment, and tenure, including a Country × Channel risk view, to identify the major contributing areas. Finally, focused on prioritization by analysing fraud contributors, risk-driver combinations, and fraud loss across different segments. The analysis highlighted significant fraud exposure across the platform, with approximately $12.56M in fraud loss and a 50.29% fraud rate in the analysed period. Ghana showed the highest fraud loss, while USSD showed the highest fraud rate among channels. Further segmentation helped identify higher-risk areas such as KYC tiers and specific country/risk combinations, providing a basis for the fraud team to prioritize investigations and corrective actions.

Skills & Tools Used:

Share this Project: