This project explores synthetic transactional data from a large e-commerce retailer to analyse customer purchasing behaviour over a 2-year period (May 2023 – May 2025)
across multiple countries. The dataset was designed with intentional inconsistencies and messiness to simulate real-world data cleaning challenges. PostgreSQL was used
for data exploration and cleaning, while Power BI was used for data modelling and visualisation.
Best Time to Sell: Morning hours lead in sales quantity
Top Category: Clothing
Forecast: Potential dip projected in mid-2025
Visualisation Summary (Power BI)
Enhanced Visibility: Offers HR leaders a clear view of key workforce metrics, enabling better understanding and management of human capital.
Data-Driven Decisions: Supports strategic HR initiatives by providing actionable insights derived from comprehensive data analysis.
Employee Insights: Helps identify trends in employee satisfaction, performance, and retention, allowing for targeted interventions and improvements.
Equity and Inclusion: Promotes fairness in compensation and gender representation, fostering a more inclusive workplace culture.
Project Timeline
Data Cleaning: PostgreSQL
Data Modelling & Visualisation: Power BI
Dashboard Creation: Power BI
Analysis & Insights: Power BI
Summary
This project demonstrates the full analytics workflow from messy raw data to actionable insights.
It showcases skills in data cleaning, normalisation, SQL, data modelling, and interactive visualisation using Power BI.
It reflects a realistic scenario of preparing data for business decision-making in e-commerce.