Scribd
Upload
13 views

Visualization

Uploaded by

Mbogo Alex
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
13 views

Visualization

Uploaded by

Mbogo Alex
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

BUSINESS INTELLIGENCE AND ANALYTICS

VISUALIZATION

BY

Mbogo Alex
Business Questions:

I will use visualization effects which are dashboards, heat maps, fever charts and dial gauges to answer
the following question:

1. To analyze and visualize the overall sales trends over time.


2. How does the average quantity ordered vary across different product lines and months?
3. To visualize the average sales value using a dial gauge.

Selected Dataset:

Sample Sales Data is the name of the dataset that includes different types of information about orders,
sales, customers, shipping, and more. Its primary aim was to facilitate segmentation, customer
analytics, clustering, and retail analytics. Initially, Pentaho Data Integration (DI) Kettle, a popular data
integration and ETL (Extract, Transform, Load) tool, was used to process the dataset. María Carina
Roldán recognized the potential for Sales Simulation training and modified it accordingly, however, as
the creator.

The dataset consists of the following columns:

1. ORDERNUMBER: A unique identifier for each order.


2. QUANTITYORDERED: The quantity of products ordered in each order.
3. PRICEEACH: The unit price of each product.
4. ORDERLINENUMBER: A sequential number assigned to each line item within an order.
5. SALES: The total sales amount for each order (calculated as QUANTITYORDERED
multiplied by PRICEEACH).
6. ORDERDATE: The date when the order was placed.
7. STATUS: The status of the order (e.g., processed, shipped, delivered, etc.).
8. QTR_ID: The quarter of the year when the order was placed (e.g., 1 for Q1, 2 for Q2, etc.).
9. MONTH_ID: The month when the order was placed (e.g., 1 for January, 2 for February, etc.).
10. YEAR_ID: The year when the order was placed.
11. PRODUCTLINE: The product line/category to which the ordered product belongs.
12. MSRP: Manufacturer's Suggested Retail Price for the product.
13. PRODUCTCODE: A unique code identifying each product.
14. CUSTOMERNAME: The name of the customer who placed the order.
15. PHONE: The contact phone number of the customer.
16. ADDRESSLINE1: The first line of the customer's address.
17. ADDRESSLINE2: The second line of the customer's address.
18. CITY: The city where the customer is located.
19. STATE: The state where the customer is located.
20. POSTALCODE: The postal code of the customer's location.
21. COUNTRY: The country where the customer is located.
22. TERRITORY: The territorial region associated with the customer's location.
23. CONTACTLASTNAME: The last name of the customer contact.
24. CONTACTFIRSTNAME: The first name of the customer contact.
25. DEALSIZE: A categorical variable indicating the size of the deal (e.g., small, medium, large).
Selected tools:

Pandas – Allows us to provide DataFrame data structures used to handle and manipulate a dataset.
Matplotlib – A visualization library for creating chrarts.
Seaborn – Python library for creating visually appealing statistical graphics.

How visualization was performed:

Dashboard:

1. Business question: To analyze and visualize the overall sales trends over time.
2. Visualization process: The data is grouped by the ‘ORDERDATE’ column and the sum of
sales for each date is calculated. The resulting data is plotted using the matplotlib library.
3. Results: The result is a line chart that depicts the sales trend over time. Stakeholders can
observe the upward or downward trends, identify peak periods and also they can assess the
overall sales trajectory.
Heat Maps:

Question: How does the average quantity ordered vary across different product lines and months?

Creating the Heatmap:

• I first created a pivot table, the pivot table calculates the average quantity ordered
(`QUANTITYORDERED`) by (`PRODUCTLINE`) and (`MONTH_ID`).
• The resulting pivot table, `heatmap_data` is the data source for the heatmap.
• The heatmap is created by passing the following parameters:
◦ z=heatmap_data.values
◦ x=heatmap_data.columns
◦ y=heatmap_data.index
◦ colorscale=‘Viridis’
Answer to the business question: The average of the quatity ordered varies across different product
lines and months as shown in the heatmap. Darker colors indicate higher average quantities ordered and
lighter colors indicate lower average quantities ordered.
Dial Gauge:

1. The business question being addressed: To visualize the average sales value using a dial
gauge.
2. ETL tool description and application: I used Python’s ‘plotly.graph_objects’ to do the
visualization.
3. Visualization process: The ‘go.Figure’ and ‘go.Indicator’ classes from ‘plotly.graph_objects’
were first initialized. Then the gauge was constructed:

4. Results: The average sales value indicated by the dial gauge was 3.55K.

You might also like