Adidas Sales Performance Analysis & Interactive Dashboard

Overview

This project provides a comprehensive analysis of Adidas sales data, aiming to extract key insights into sales performance across different dimensions such as retailers, product categories, regions, and time. It showcases a typical data analysis workflow, including data loading, cleaning, aggregation, and visualization using Python's Pandas, Matplotlib, and Plotly libraries. Furthermore, an interactive dashboard built with Streamlit (app.py) is included for dynamic exploration of the findings, allowing users to interact with visualisations and filter data in real-time. The core data processing and static visualisations are documented within a Jupyter Notebook (main.ipynb).

Key Questions / Objectives

• Analyze total sales performance by different retailers, visualising the contribution of each.
• Understand sales trends over time by tracking monthly sales figures.
• Examine the distribution of total sales and units sold across various states, using dual-axis charts for comparative insights.
• Explore sales breakdown hierarchically by region and city using interactive treemaps, allowing drill-down into geographical sales performance.
• Identify top-performing product categories and their impact on overall sales.
• Provide a user-friendly interactive dashboard to facilitate easy data exploration and insight discovery.

Project Structure

File Name	Description
main.ipynb	The primary Jupyter Notebook containing the step-by-step data loading, cleaning, aggregation, and initial visualisation using Pandas and Plotly. This is where the in-depth analytical process is documented.
app.py	A Python script for the interactive Streamlit web application. This script reads the Adidas sales data and generates the dynamic dashboard visualisations.
data/Adidas.xlsx	The raw sales data in Excel format, which serves as the input for both the Jupyter Notebook and the Streamlit application.
README.md	This README file provides an overview, setup instructions, and details about the project.
images/logo.jpeg	An image file specifically used within the Streamlit application (app.py) for branding or visual elements.

Technologies / Libraries Used

• Python 3.x
• Pandas: For efficient data loading, manipulation, and advanced analytical operations on tabular data.
• NumPy: For fundamental numerical computing (often used implicitly by Pandas).
• Matplotlib: For foundational plotting capabilities, though more complex interactive plots are handled by Plotly.
• Plotly Express & Plotly Graph Objects: Utilised extensively for creating rich, interactive, and visually appealing data visualisations such as bar charts, line charts, dual-axis charts, and treemaps.
• Streamlit: The framework used to transform the data analysis and visualisations into an interactive web application, making the insights accessible and explorable.
• Openpyxl: To enable Pandas to read and process .xlsx Excel files.

Setup and Installation

1. Clone the Repository

   git clone [https://github.com/juto-shogan/adidas-Analysis.git](https://github.com/juto-shogan/adidas-Analysis.git)
   cd adidas-Analysis

2. Create a Virtual Environment (Recommended)

   python -m venv venv
   # On Windows:
   # .\venv\Scripts\activate
   # On macOS/Linux:
   # source venv/bin/activate

3. Install Dependencies

   Create a requirements.txt file in your project root with the following contents:

   pandas
   matplotlib
   plotly
   streamlit
   openpyxl
   

   Then install them using pip:

   pip install -r requirements.txt

How to Run the Analysis & Dashboard

1. Activate your virtual environment (if created):

   • On Windows: .\venv\Scripts\activate
   • On macOS/Linux: source venv/bin/activate

2. Explore the Jupyter Notebook (main.ipynb): To delve into the detailed analysis process, data transformations, and the generation of individual plots:

   jupyter notebook main.ipynb

   This command will open the Jupyter Notebook in your default web browser, where you can execute cells sequentially to review and reproduce the data analysis steps.

3. Run the Interactive Dashboard (app.py): To launch the interactive Streamlit dashboard and explore the sales insights dynamically:

   streamlit run app.py

   This command will open the interactive dashboard in your default web browser, typically at http://localhost:8501.

Findings / Insights

• Retailer Performance: The analysis identifies top-performing retailers by total sales, providing a precise understanding of which sales channels are most effective.
• Temporal Sales Trends: Monthly sales trends are visualised, revealing seasonal patterns, growth, or decline periods in Adidas sales over time.
• Geographical Sales Distribution: A combined visualisation of total sales and units sold by state allows for a nuanced comparison, highlighting areas with high revenue generation versus high product movement.
• Regional and City-Level Sales Breakdown: Interactive treemaps offer a hierarchical view of sales performance across different regions and cities, enabling drill-down into specific geographical areas for detailed insights.
• [Add any other specific insights you gained, e.g., "Discovered a significant increase in online sales during specific promotional periods," "Identified a particular product line contributing disproportionately to profits." ]

Author

Somto Mbonu

Data Analyst GitHub Profile: juto-shogan