This project consists of 6 tables with operational and sales data for a Brazilian e-Commerce company A&B, similar to Amazon. Data loading, exploration, cleaning, analysis and visualization was all conducted in Power BI.
2. Introduction
Microsoft Power BI is a powerful tool known for its dynamic visualization capabilities,
yet it offers so much more than that. As an example of its versatile components, I have
completed transformation, exploration, analysis and visualization of the following
dataset within Power BI.
This dataset provides powerful insights into sales and operations for e-Commerce
company A&B (similar to Amazon). The presentation is intended to summarize the
most relevant informational and actionable items for the company’s executives.
3. DATASET
• Tables—Dataset is composed of 6 different tables with multiple variables each:
• Orders dataset: Provide information for each item ordered
• Order items dataset: Information for items within each order and the cost to ship and price
broken out for each item within an order.
• Order payments dataset: Provides information regarding payments made on each order.
Make sure you aggregate total payment for each order to get unique order price.
• Product dataset: Provides information about the product.
• Product category name translated dataset: This dataset is Brazilian, so all categories are in
Portuguese, join category name on this table to get the translated category.
• Order reviews dataset: This table has review information for each order.
• Customers dataset dataset: This dataset has information regarding customer_id, which links
directly to order_id in the orders dataset. However, to get the unique customer id for each
order you need to link to this table.
https://www.kaggle.com/datasets/olistbr/brazilian-ecommerce/discussion/71650
5. Revenues, Orders and Payment Types
• The data suggests that both
revenues and total orders peaked in
2017 Q4 and 2018 Q1/Q2, then
started to decline right before the
start of Q3.
• This would suggest that sales
may be seasonal and
increasing sales/promotions
during slow seasons may be
suitable.
• It’s important to recognize that
data for Q3 may also be
incomplete.
• Most common form of payment
appears to be credit card.
• Forming partnerships with
credit providers may result
beneficial for the company, as
well as the customers.
6. Revenues and Orders by States
• Data suggests that the top 3 states
with the highest revenue and total
number of orders are SP, RJ and
MG.
• This information would help
A&B make informed decisions
on where to focus their
marketing/sales efforts.
7. Orders and Profits by Categories
• Top 10 most profitable categories are
almost the same as the top 10 most
popular.
• However, there are some minor
differences between the two.
• Toys and telephony are
on the top 10 most
common, but not on the
top 10 most profitable.
• Conversely, garden tools
is among the most
profitable. Thus, it would
be worth devoting more
marketing efforts into
this category.
8. Customer Satisfaction
• This chart shows the relation
between “Average Review Score” and
the total number of orders .
• Highest total number of orders
appears to be in the mid-to-low
average review scores.
• This would motivate managers
to become familiar with the
major customer pain points
based on reviews.
10. Data Acquisition, Preparation and Analysis
Microsoft Power BI
• Data was acquired via Kaggle and downloaded into Excel Files.
• Cleaning, transformation, analysis and visualization of the data was ALL conducted in Power BI.
• Due to the multiple tables of data, a data model was created, and multiple joins were conducted resulting in a final merged
table.
11. Data Acquisition, Preparation and Analysis
Microsoft Power BI
• Within Power Query, multiple tables were joined.
• Data was also clean, standardized and grouped to reduce cardinality.
• Location data types were changes to represent geographical points in order to create maps.
• Given the size of the dataset, unnecessary columns were removed, and new custom columns were added.
• Column distribution, quality and profile tools were used to analyze the content of the columns.
12. Data Analysis and Visualization
Microsoft Power BI
• To analyze and visualize the data, multiple groups, charts, matrix filters and slicers were used.
• Dashboard-wide
filters were
used to exclude
transactions
before 2017.
• Page wide
filters were also
used to meet
the needs of
each visual.
• Categories were
grouped to
minimize
cardinality.
• Slicers were
integrated to
provide the
audience a more
dynamic
experience.
14. Challenges and Cool Techniques
• Multiple tables
• The dataset consisted of 6 main tables that had to be joined in order to analyze and visualize the information.
• Cool Technique
• Power BI provided a very helpful model view that allows the user to outline the relationship between tables.
• High Cardinality
• The dataset consisted of 6 main tables that had to be joined in order to analyze and visualize the information.
• Cool Technique
• Grouping categories is another genius feature within Power BI, along with the column profile which allows you to get
a sense of what data may need to be standardized and/or grouped.
What If I had More Time?
• If I had more time, I would look into repeat customers and some of the categories that may be sold/marketed together
based on consumer trends.
15. THE END!
Thank you for checking out my project!
If you have any feedback, comments, ideas, JOB offers, please feel free to
DM me at my LinkedIn @https://www.linkedin.com/in/ximenab/