Exploratory Data Analysis (EDA) – Global Superstore
Overview This project performs an in-depth Exploratory Data Analysis (EDA) on the Global Superstore dataset to uncover patterns in sales, profit, and customer behaviour. The analysis focuses on identifying key business insights that can support data-driven decision-making in a retail context.
Objectives Understand overall sales and profit performance Identify top-performing products and categories Analyse regional and segment-level performance Detect trends, patterns, and potential inefficiencies
Tech Stack Python Pandas & NumPy – data manipulation Matplotlib & Seaborn – data visualisation
Dataset Global Superstore dataset
Includes: Orders and sales data Product categories and sub-categories Customer segments Regional and geographical information
Project Workflow
- Data Exploration Reviewed dataset structure, columns, and data types Generated summary statistics for key variables
- Data Cleaning Checked for missing values and inconsistencies Prepared dataset for analysis
- Sales & Profit Analysis Analysed total sales and profit performance Compared profitability across categories and sub-categories
- Regional Analysis Evaluated sales performance across regions Identified high-performing and underperforming areas
- Customer Segment Analysis Compared sales and profit across customer segments Identified the most valuable customer groups
- Visualisation Created charts to highlight trends and patterns Used visual storytelling to communicate insights
Key Insights Certain product categories generate high sales but lower profit, indicating potential cost inefficiencies Regional performance varies significantly, highlighting opportunities for targeted strategies Specific customer segments contribute disproportionately to overall revenue Some sub-categories consistently underperform, suggesting potential for optimisation or removal
Limitations Analysis is based on historical data and may not reflect future trends Limited external factors (e.g., economic conditions, competition) Dataset scope may not capture all business variables
Future Improvements Build predictive models for sales forecasting Integrate additional datasets for deeper insights Develop interactive dashboards (Power BI / Tableau) Perform customer segmentation using clustering techniques
Portfolio Value
This project demonstrates:
Strong EDA and data analysis skills Ability to extract business insights from data Proficiency in data visualisation and storytelling Understanding of retail and sales analytics