This project analyzes product sales data for 2019. The dataset consists of monthly sales reports that have been merged and cleaned for analysis. The project aims to gain insights into sales trends, customer purchasing behavior, and product performance.
The business wanted to understand its sales trends, peak sales period, and customer buying behavior to improve marketing and inventory decisions
- The dataset contains sales data for 12 months of 2019.
- The data includes Order ID, Product, Quantity Ordered, Price Each, Order Date, and Purchase Address.
My goal was to analyze data to identify
1) The best Month and City for sale
2) The Optimal time for Advertisement
3) Frequently bought products together
4) Top-selling products with a reason for their performance
I used Python throughout the project. To merge the monthly files, I used the os library, then performed all the data cleaning and transformations with pandas — like extracting the month, hour, and city. For calculating totals and other stats, I used NumPy. And to present insights visually, I built charts using matplotlib and seaborn.
While working on the project, I faced a few challenges, like merging large files efficiently and visualizing some complex patterns. So, I referred to documentation, used Google, and even took some help from ChatGPT to understand certain techniques better. It helped me speed up my learning and apply best practices.
As a result of the analysis, I found that December had the highest sales, around 4.6M, likely due to holiday shopping. San Francisco had the most orders, and sales peaked around Afternoon 11 AM, 12 AM, and Evening 7 PM, suggesting those are ideal times for running ads. I also found that batteries and USB-C cables were frequently bought together, so recommending bundles could increase revenue. These insights could directly help with optimizing marketing and stock management.