Skip to content

kelechiu/Exploratory-Data-Analysis-using-Python

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

E-Commerce Sales Analysis using python

Introduction

Superstore’s products are grouped into Furniture, Office Supplies and Technology while its customers are either Consumer, Corporate or Home Office. The aim of this project is to provide an in-depth analysis of this data and to answer the following questions:

  • What sub-categories generate the most/least revenue and profit?

  • What products are the most/least profitable?

  • What states generate the most sales and profit?

  • What are our customers' purchasing habits?

Project Strategy

The superstore dataset was downloaded from Kaggle and the Python libraries used on this project were Pandas, Matplotlib and Seaborn. The dashboard was designed with Tableau and Figma.

The main steps for this project are summarized below:

  • Data Preparation/Cleaning
  • Exploratory Data Analysis
  • Insights
  • Recommendations

Insights

1) What (Sub)Categories generate the most/least revenue and profit?

  • Phones and Chairs generated the most revenue, both accounting for 29% of the Total Sales. Fasteners and Labels generated the least revenue.

  • The least profitable sub-category is Tables with a net loss of ($18,000). It’s important to note that loss-making sales like Tables, Bookcases and Supplies make up 16% of total sales.

  • 60% of all orders are for Office Supplies and 26% of those are for Binders.

2) What products are the most/least profitable?

  • Canon ImageClass Copier is the best-selling and most profitable product while Cubify CubeX 3D Printers and Lexmark Laser Printer are the least profitable.

3) What states generate most/least sales and profit?

  • California and New York had the highest sales and profit, accounting for 64% of Superstore’s total sales and 66% of Superstore’s total profit.

  • The states with the least profit are North Dakota and West Virginia.

4) What are our customers' purchasing habits?

For the purpose of this analysis, I will refer to Active Customers as those that have placed orders in the last 30 days.

  • Of all 693 total customers, 75% have ordered at least 5 times and 11% have ordered more than 10 times

  • Only 23% of customers are active and have made purchases in the last 30 days

  • A total of 55 new customers were gained in the past year

  • Of all 5009 total orders, only 12 are one-off orders

  • 60% of all orders are Office Supplies

5) Recommendations

  • Create a Home Office package with both high-selling categories like Phones and Chairs and less profitable products like Tables and Bookcases to offset losses.

  • Consider increasing the price of Tables as it’s a high-selling product. However, Bookcases and Supplies could be either dropped from the product list or negotiated with suppliers for a cheaper price.

  • Customer Success teams should target inactive customers to drive sales and marketing goals. Special discounts on their favorite products can be offered to retain the relationship and prevent churn.

Dashboard 1 (2)

Click here to read the full analytics report

About

A repository for e-commerce sales analysis using python

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published