4 of 37

EDA

Exploratory Data Analysis refers to the critical process of performing initial investigations on data so as to discover patterns, to spot anomalies, to test hypothesis and to check assumptions with the help of summary statistics and graphical representations.
EDA explained using sample dataset: To share our understanding of the concept and techniques we know, we take an example of white variant of Brazilian E-Commerce Public Dataset by Olist which is available on Kaggle and try to catch hold of as many insights from the data set using EDA.

5 of 37

About Data

This Data was taken from a website named Kaggle.com
The Data is about an E-commerce store in Brazil stretching over the span of almost 2 Years
The data tell us about different categories’ products, sales, revenue and number of orders
It also tells us about states, cities and sellers
We have 8 tables in this dataset including more than Hundred thousands

7 of 37

Total Revenue share by State

We have identified, in this analysis, which state has the largest customer market in term of revenue.

We can see through the charts that ‘SP’ state hold the largest share in the market with almost 6 Million of revenue and approximately 43% of the entire market.

This analysis can help the business to find their potential customer and help them create business strategies in the states where there is less or no market.

8 of 37

Total Revenue share by State

9 of 37

Top 10 Categories with their best sellers

We have identified, in this analysis, top 10 categories based on revenue generated by the sellers.

We have also found the best seller and their revenue in those categories.

We can see through the chart that ‘Beleza Saude’ has generated the highest amount of revenue of approximately 1.4 Million with the best seller generating approx. 83k.

This analysis can help the business to find their best categories and find the best sellers of the category and also the sellers that are liability.

10 of 37

Top 10 Categories with their best sellers

11 of 37

Top 10 cities based on revenue and their best seller’s revenue

We have identified, in this analysis, top 10 cities based on revenue generated by the sellers.

We have also found the best seller and their revenue in those cities.

We can see through the chart that Sao Paulo has generated the highest amount of revenue of approximately 3.1 Million with the best seller generating approx. 172k

This analysis can help the business to find their best cities and find the best sellers of the city and the sellers that are liability

12 of 37

Top 10 cities based on revenue and their best seller’s revenue

13 of 37

Revenue by Months

We have found revenue generated in each months with one month missing.

This analysis can help the business to find the increase and decrease in the revenue each month throughout the period and which months are busy so that they can take measures according to it and predict their future revenue.

14 of 37

Revenue by Months

15 of 37

Most Payment method used by customer

In this analysis, we have found which payment method is more popular among the buyers.

We can see that the credit card is mostly used by the customer, approximately ¾ times.

This can help the company to create ease for the credit card holders by taking different measures so that they can create better user experience for the customers.

16 of 37

Most Payment method used by customer

17 of 37

Average review of Top 10 category

By this analysis we can see the Top 10 categories reviewed by the customer.

This can help us to improve less review categories.

18 of 37

Average review of Top 10 category

19 of 37

Average Review vs Delivery

This analysis show the customer’s reviews depends upon order delivery time.

We can clearly see that the increase in delivery time decreases the customer feedback.

The analysis can help the company to take compulsory measures to decrease the delivery time and better customer feedback.

20 of 37

Average Review vs Delivery

21 of 37

Top 10 categories based on order count

In this analysis, we have found the top 10 categories generating highest amount of orders

It can clearly seen through the chart ‘Bed Bath Table’ contains the most amount of orders

22 of 37

Top 10 categories based on order count

23 of 37

�Top 10 sellers based on revenue generated

In this analysis we have found which seller has generated the highest amount of revenue overall.

This can help the business identify their top seller’s performance and rate them accordingly

24 of 37

Top 10 sellers based on revenue generated

25 of 37

Top 10 products based on units sold and their revenue

In this analysis, we have found the top 10 products based on units sold of each product and the revenue generated by those products

This will help the business find which products need betterment and which need to be removed

26 of 37

Top 10 products based on units sold and their revenue

27 of 37

Late and on-time Delivered orders

We found, in this analysis, the order delivered within the estimated time and beyond the estimated time

We can clearly see through the chart approx. 92%(90k approx.) orders have been delivered within the estimated time successfully

28 of 37

Late and on-time Delivered orders

29 of 37

TOTAL ORDERS PER STATUS

IN THIS ANALYSIS, WE HAVE FOUND THE ORDER COUNT ON DIFFERENT STATUSES THAT ARE: DELIVERED, INVOICED, SHIPPED, PROCESSING, CANCEL, CREATED, APPROVED

WE CAN CLEARLY SEE THAT 97% OF THE TOTAL ORDERS ARE DELIVERED

30 of 37

TOTAL ORDERS PER STATUS

31 of 37

Purchasing trend of customers in month and years

We have found, in this analysis, the numbers of orders generated in each month of each year.

We can see the trends followed by each year.

1 of 37

2 of 37

3 of 37