Description
This project performs an in-depth analysis of Amazon product data. The objective is to draw insights regarding product ratings, sales, categories, and pricing. Through this analysis, we aim to understand consumer behavior, popular product categories, and the relationship between product price and customer satisfaction.
The dataset used in this project contains information on various Amazon products including their prices, ratings, categories, and number of sales.
Key Findings
The most sold and popular items on Amazon are Smart Watches and USB Cables. The highest-rated products were mostly electronic devices and accessories. The most expensive products are mostly electronics, especially Smart TVs and High-End Smartphones. The category with the most products listed is USB Cables, followed by Smart Watches and Smartphones. The highest revenue generating categories are Televisions and Smartphones. A mild correlation was found between price and ratings, indicating that high price doesn't necessarily mean high customer satisfaction. The categories with the highest average prices were air conditioners and laptops.
Files
Workflow - representation of the workflow used in this project.
The project consists of: amazon.csv - This is the dataset file used for the analysis. It is a CSV file containing details about various Amazon products.
wrangle-ing.xlsx - contains the csv data and includes the conversion from the Indian Rupee to the United States Dollar, that was used for the analysis of this project.
sort_data.ipynb - this contains the sorting of the data to answer the 7 business intelligence questions that we will look at. files from sort_data.ipynb = sorted_data.py; az_refined.csv; az_refined.xlsx
Business_Intelligence_Analysis_of_Amazons_Sales_Data.ipynb - This is the Jupyter notebook containing the analysis and visualizations.
Instructions
To run the notebook, you will need Python along with the following libraries installed on your machine:
Pandas Numpy Matplotlib Seaborn Once you've installed the necessary libraries, clone this repository and run Jupyter notebook. You can then navigate to Business_Intelligence_Analysis_of_Amazons_Sales_Data.ipynb to view the analysis.
Future Work
Future improvements and extensions to this project could include time series analysis to understand sales trends over time, a comparison of product performance across different regions, and predictive modeling to forecast future sales.
Author
dres_code
Acknowledgements
This project is based on data obtained from Amazon, and is intended for educational purposes only. All trademarks and registered trademarks are the property of their respective owners.