Answer To: Page 1 of 4 CRICOS Provider No. 00103D ITECH 1103 Group Report Semester 3 2019 ITECH1103- Big Data...
Pooja answered on Jan 29 2021
analysis of customer data
Table of Contents
Executive Summary 4
Background information 5
Analysis 6
1) Data dictionary 6
2) Number of Customers 6
3) Customers and countries 6
4) Country and Revenue 7
5) Customers and total revenue 8
6) days to place and order 9
7) Customer and quantity ordered 10
8) Customer type and quantity ordered 11
9) top 5 countries with respect to each customer type 12
10) customer type and continent by quantity ordered 13
11) Title 14
12) Days of Delivery 14
13) Most efficient predictor of Quantity ordered 15
14) country of customer 16
15) Cluster Analysis for Quantity ordered on basis of Customer Type 16
16) Days to delivery for each continent type by customer type. 16
17) days to delivery by customer type name 18
18) days to deliver & number of orders by order type 18
19) cost, days to deliver and discount percent 19
20) customer type with unit cost and quantity ordered 19
Discussion of findings 21
2) Number of Customers 21
3) Customers and countries 21
4) Country and Revenue 21
5) Customers and total revenue 21
6) days to place an order 21
7) Customer and quantity ordered 22
8) Customer type and quantity ordered 22
9) top 5 countries with respect to each customer type 22
10) customer type and continent by the quantity ordered 22
11) Title 23
12) Days of Delivery 23
13) Most efficient predictor of Quantity ordered 23
14) country of the customer 23
15) Cluster Analysis for Quantity ordered on basis of Customer Type 23
16) Days to delivery for each continent type by customer type. 23
17) days to delivery by customer type name 23
18) days to deliver & number of orders by order type 24
19) cost, days to deliver and discount percent 24
20) customer type with the unit cost and quantity ordered 24
Other Visualizations 25
Profit 25
Profit and continent 25
Conclusion 27
References 29
Figure 1 5
Figure 2 6
Figure 3 7
Figure 4 8
Figure 5 9
Figure 6 10
Figure 7 11
Figure 8 12
Figure 9 13
Figure 10 13
Figure 11 14
Figure 12 15
Figure 13 15
Figure 14 16
Figure 15 17
Figure 16 17
Figure 17 18
Figure 18 18
Figure 19 19
Table 1 5
Table 2 6
Table 3 6
Table 4 8
Table 5 9
Table 6 9
Table 7 10
Table 8 11
Table 9 12
Table 10 13
Table 11 14
Table 12 17
Executive Summary
The total number of customers is 9,51,669. The top 10 customers considering all countries correspond to customers Antonio, Barbara, Andrea, Alain, Andrew, Ana, Anne, Andreas, Alberto, Anna. The top 10 countries with the highest total of the retail price are the United States, Germany, France, Italy, United Kingdom, Spain, Netherlands, Australia, Belgium, Denmark. Michael and David have a retail price of $8,79,353 and $8,24,643 respectively. The highest number of orders were placed on 28th February 2014, and 19th December 2016. The highest quantity ordered was by customers David, Michael, and John with a value of 10600, 10192, and 10000. The customer type Orion Club Gold members high activity, Orion Club Gold members medium activity, and Orion Club members high activity have the highest Quantity Ordered. For all 5 types of customers, the United States and Germany are included in the top 5 countries. The Orion Club Gold member's high activity gave 258788 orders to Europe. The cities Madrid, London, Paris, and Milano have maximum days of delivery. Cost appears to be the most important predictor of the quantity ordered. The maximum number of customers belongs to the United States. There are 4 clusters created for Customer Type considering Quantity ordered. The days to delivery are the highest by Europe for customer type Internet/Catalog Customers with a total of 161816 days. The top 3 customer types receiving the fastest delivery are Orion Club members low activity, Orion Club Gold members low activity, and Orion Club members medium activity.The maximum number of order types corresponds to retail sales. There is a weak linear relationship between (cost, days to deliver) and (discount percentages, days to deliver). There is a strong positive linear relationship between cost and discount percentage.
Background information
The customer segmentation provides insight into the landscape of the market revealing customer characteristics that can be used to group customers into segments that have something in common. this process is also known as clustering and the techniques used to develop these models are called clustering algorithms.
Visualizations to customer data is applied to give the depth of discovery. Visualizations give basic trends like predictions questions like What if queries and you can adjust, everything hypothetically and visualize all the components of data dynamically for different comparisons.
Analysis
1) Data dictionary
The dataset has 22 columns or variables involved. The categorical variables are City Name, Continent Name, Customer Birth Date, Customer Country, Customer Group Name, Customer ID, Customer Type Name, Customer First name, Customer Last name, Date order was delivered, date order was placed, Order ID, Order Type, Postal code, State name, and Title. The continuous variables are Cost, Days to delivery, Discount in percent of Normal total retail price, Frequency, Profit, Quantity ordered, Retail price, and Frequency percent.
2) Number of Customers
Table 1
Customer Group Name
Frequency
Internet/Catalog Customers
76,965
Orion Club Gold members
4,83,438
Orion Club members
3,91,266
Count: 9,51,669
Figure 1
3) Customers and countries
Table 2
Customer_FirstName
Frequency
Antonio
2,581
Barbara
2,480
Andrea
2,329
Alain
1,921
Andrew
1,888
Ana
1,731
Anne
1,490
Andreas
1,485
Alberto
1,433
Anna
1,380
Angela
1,075
Alan
1,065
Figure 2
Figure 3
4) Country and Revenue
Table 3
Customer Country
Retail Price
$13,31,96,488.48
United States
$3,29,01,896.80
Germany
$1,97,60,678.03
France
$1,47,01,399.68
Italy
$1,45,25,623.82
United Kingdom
$1,31,75,549.39
Spain
$1,22,33,250.16
Netherlands
$95,49,237.18
Australia
$78,94,103.00
Belgium
$31,37,551.64
Denmark
$24,62,872.19
Figure 4
5) Customers and total revenue
Table 4
Customer_FirstName
Retail Price
Michael
$8,79,353.92
David
$8,24,643.56
John
$7,94,653.59
Robert
$6,94,692.47
Peter
$6,48,676.25
James
$5,76,694.37
Thomas
$5,71,126.76
Paul
$4,89,962.85
William
$4,49,091.91
Christine
$4,12,559.50
Figure 5
6) days to place an order
Table 5
Date Order was placed by Customer
Frequency
February 28, 2014
1,096
December 19, 2016
1,003
December 06, 2016
998
December 13, 2016
931
December 27, 2016
926
December 12, 2016
914
December 26, 2016
912
December 20, 2016
899
December 05, 2016
884
December 16, 2016
876
December 02, 2016
872
November 01, 2016
857
August 09, 2016
855
July 26, 2016
854
December 08, 2016
853
December 18, 2016
849
July 08, 2014
843
December 01, 2016
839
December 15, 2016
836
December 03, 2016
835
Figure 6
7) Customer and quantity ordered
Table...