Heading
Revenue Loss Identification
Year
2025
Type of Project
Side Project
My Role
Data Analyst
Case Study
Objective
The goal was to quantify revenue leakage from misapplied discounts and pricing errors, determine root causes and prescribe operational and technical controls to prevent recurrence. Methodology combined SQL extracts, Python-based validation and manual audit sampling.
Process
Sourced a synthetic dataset generated through gemini.ai, simulating real world distributions. Utilised SQL for extraction & views, Python (pandas) for cleansing & validation and Power BI for visualisation. Used historical transactions (12 months), retail and channel sales and all discount types applied at checkout.
Outcome
126 transactions exceeded the allowable discount cap of 1.5%, representing widespread exceptions rather than isolated incidents. These transactions spanned multiple product categories and sales channels, indicating the possibility of system gaps.
Standout Features
Configuration errors.
Authorisation Gaps.
System Integration Issues.