MRA PROJECT

MILESTONE 1
CONTENTS

EXPLORATORY DATA ANALYSIS

 Problem Statement & Data Dictionary
 Data Description
 Univariate Analysis (Numerical & Categorical Variables)
 Bivariate Analysis
 Multivariate Analysis
 Sales Trend
 Sales across different Categories
 Inferences
CONTENTS

RFM ANALYSIS – CUSTOMER SEGMENTATION

 RFM
 KNIME - Workflow
 Output Table & Segments
 Segmentation Output (Inferences)
 Best Customers
 Customers on the verge of churning
 Lost Customers
 Loyal Customers
PROBLEM STATEMENT

An automobile parts manufacturing company has collected data of transactions for 3 years.
They do not have any in-house data science team, thus they have hired you as their consultant.
Your job is to use your magical data science skills to provide them with suitable insights about their data
and their customers
 EDA

EXPLORATORY DATA ANALYSIS

Data Dictionary
ORDERNUMBER : Order Number
QUANTITYORDERED : Quantity ordered
PRICEEACH : Price of Each item
ORDERLINENUMBER : order line
SALES : Sales amount
ORDERDATE : Order Date
DAYS_SINCE_LASTORDER
Days_
: Since_Lastorder
STATUS : Status of order like Shipped or not
PRODUCTLINE : Product line – CATEGORY
MSRP : Manufacturer's Suggested Retail Price
PRODUCTCODE : Code of Product
CUSTOMERNAME : customer
PHONE : Phone of the customer
ADDRESSLINE1 : Address of customer
CITY : City of customer
POSTALCODE : Postal Code of customer
COUNTRY : Country customer
CONTACTLASTNAME : Contact person customer
CONTACTFIRSTNAME : Contact person customer
DEALSIZE : Size of the deal based on Quantity and Item Price
 EDA
DATA DESCRIPTION

 The Data set is about the Sales of an automobile part manufacturer.

 To find : The underlying buying patterns of the customers
 Data is of 2747 rows and 20 columns
 There are 0 null entries and 0 duplicates
 There are 4 data types : Int, Float, Object and DateTime
 EDA

DATA INFO & SUMMARY STATISTICS

Data Info Data Summary Statistics

 EDA
UNIVARIATE ANALYSIS – NUMERICAL VARIABLES
EDA
EDA
 EDA

UNIVARIATE ANALYSIS – CATEGORICAL VARIABLES

EDA
EDA
 EDA

BIVARIATE ANALYSIS
EDA
EDA
EDA
EDA
EDA
EDA
EDA
 EDA

MULTIVARIATE ANALYSIS
EDA
EDA
EDA
EDA
EDA
EDA
EDA
EDA
EDA
 EDA

INFERENCES - EDA

 Outliers are present in QuantityOrdered, PriceEach, Sales and MSRP

 Out of all the Productline, Classic Cars has the highest count, Trains has the lowest count
 Out of all the Orders, most number of orders are under Shipped status and least are Disputed
 USA is the country with most orders, Ireland is the country with least orders
 Deal size – Medium has the most orders, Large has the least orders
 Sales across different Categories
 Medium Deal size has highest Sales and Large has the least Sales
 Madrid city has the highest Sales
 USA has the highest Sales and Ireland has least Sales
 EDA

INFERENCES - EDA

 The sales of large size deal is almost remain stagnant over the years and it can be presumed that
company should focus on getting large size chunk projects
 The company is customer driven because there major chunk of sales comes from 4 – 5 customer. So therefore
company should focus more on customer scouting in a rationale way because in case there is client churn's it will
impact the sales of the company grossly
 The sales of product line like vintage cars and motorcycles is in saturation stage and there is no scope in them,
however trucks and buses product line is still have scoop for sales
 Company drives export revenue mainly from large size deals followed by medium size deals and small size deals
therefore it can be inferenced that large size deals is mainly from foreign customers and company should focus
on domestic sales as well
 RFM
RFM ANALYSIS

 RFM Analysis is a marketing technique used to quantitatively

rank and group customers based on the recency, frequency
and monetary total of their recent transactions to identify
the best customers and perform targeted marketing
campaigns
 The Recency, Frequency, Monetary Value (RFM)
Model assigns a firm's customer base a particular trait,
which can be used to improve marketing analysis
 RFM analysis enables marketers to increase revenue by
targeting specific groups of existing customers (i.e.,
customer segmentation) with messages and offers that are
more likely to be relevant based on data about a particular
set of behaviour
 RFM
RFM ANALYSIS – WORKFLOW

 RFM Analysis is done using KNIME tool

 Recency analysis is done using today’s
date as the reference point (15/10/2022)
 Frequency analysis is done using the
Order Number
 Monetary analysis is done using
Price_Each and Quantity_Ordered
 RFM

RFM ANALYSIS - OUTPUT

 This is the first 5 rows of the output of KNIME workflow shown in the previous slide.
CUSTOMERNAME ORDERNUMBER QUANTITYORDERED MONETARY RECENCY RECENCY_SEGMENTS MONETARY_SEGMENTS FREQUENCY_SEGMENTS
AV Stores, Co. 51 34.8627451 157807.81 1063 Less_Recent High Frequent
Alpha Cognac 20 34.35 70488.44 931 Most_Recent Low Lost
Amica Models & Co. 26 32.42307692 94117.26 1132 Least_Recent Medium Rare
Anna's Decorations, Ltd 46 31.93478261 153996.13 950 Recent High Frequent
Atelier graphique 7 38.57142857 24179.96 1055 Less_Recent Low Lost

 The segments used for each category from RFM are as below

Percentile / Category Recency Frequency Monetary

0.00 - 0.25 Most_Recent Lost Low
0.25 - 0.50 Recent Rare Average
0.50 - 0.75 Less_Recent Occasional Medium
0.75 - 1.00 Least_Recent Frequent High
 RFM
RFM ANALYSIS – SEGMENTATION OUTPUT

 BEST CUSTOMERS

CUSTOMERNAME ORDERNUMBER QUANTITYORDERED MONETARY RECENCY RECENCY_SEGMENTS MONETARY_SEGMENTS FREQUENCY_SEGMENTS

Danish Wholesale Imports 36 36.52777778 145041.6 913 Most_Recent High Frequent
Euro Shopping Channel 259 36.01158301 912294.11 867 Most_Recent High Frequent
L'ordine Souveniers 39 32.82051282 142601.33 888 Most_Recent High Frequent
La Rochelle Gifts 53 34.56603774 180124.9 867 Most_Recent High Frequent
Mini Gifts Distributors Ltd. 180 35.36666667 654858.06 869 Most_Recent High Frequent
Reims Collectables 41 34.95121951 135042.94 929 Most_Recent High Frequent
Salzburg Collectables 40 36.05 149798.63 881 Most_Recent High Frequent
Souveniers And Things Co. 46 34.80434783 151570.98 869 Most_Recent High Frequent
The Sharp Gifts Warehouse 40 41.4 160010.27 906 Most_Recent High Frequent
 RFM
RFM ANALYSIS – SEGMENTATION OUTPUT

 CUSTOMERS ON THE VERGE OF CHURNING

CUSTOMERNAME ORDERNUMBER QUANTITYORDERED MONETARY RECENCY RECENCY_SEGMENTS MONETARY_SEGMENTS FREQUENCY_SEGMENTS

Auto Assoc. & Cie. 18 35.38888889 64834.32 1100 Least_Recent Low Lost
Bavarian Collectables Imports, Co. 14 28.64285714 34993.92 1126 Least_Recent Low Lost
CAF Imports 13 36 49642.05 1306 Least_Recent Low Lost
Cambridge Collectables Co. 11 32.45454545 36163.62 1256 Least_Recent Low Lost
Clover Collections, Co. 16 30.625 57756.43 1125 Least_Recent Low Lost
Daedalus Designs Imports 20 34.95 69052.41 1332 Least_Recent Low Lost
Double Decker Gift Stores, Ltd 12 29.75 36019.04 1362 Least_Recent Low Lost
Iberia Gift Imports, Corp. 15 39.26666667 54723.62 1105 Least_Recent Low Lost
Online Mini Collectables 15 38.13333333 57197.96 1131 Least_Recent Low Lost
Osaka Souveniers Co. 20 34.6 67605.07 1281 Least_Recent Low Lost
Signal Collectibles Ltd. 15 34.26666667 50218.51 1343 Least_Recent Low Lost
West Coast Collectables Co. 13 39.30769231 46084.64 1355 Least_Recent Low Lost
 RFM
RFM ANALYSIS – SEGMENTATION OUTPUT

 LOST CUSTOMERS

CUSTOMERNAME ORDERNUMBER QUANTITYORDERED MONETARY RECENCY RECENCY_SEGMENTS MONETARY_SEGMENTS FREQUENCY_SEGMENTS

Alpha Cognac 20 34.35 70488.44 931 Most_Recent Low Lost
Atelier graphique 7 38.57142857 24179.96 1055 Less_Recent Low Lost
Australian Gift Network, Co 15 36.33333333 59469.12 986 Recent Low Lost
Auto Assoc. & Cie. 18 35.38888889 64834.32 1100 Least_Recent Low Lost
Auto-Moto Classics Inc. 8 35.875 26479.26 1047 Recent Low Lost
Bavarian Collectables Imports, Co. 14 28.64285714 34993.92 1126 Least_Recent Low Lost
Boards & Toys Co. 3 34 9129.35 980 Recent Low Lost
CAF Imports 13 36 49642.05 1306 Least_Recent Low Lost
Cambridge Collectables Co. 11 32.45454545 36163.62 1256 Least_Recent Low Lost
Classic Legends Inc. 20 36 77795.2 1059 Less_Recent Average Lost

 LOYAL CUSTOMERS
CUSTOMERNAME ORDERNUMBER QUANTITYORDERED MONETARY RECENCY RECENCY_SEGMENTS MONETARY_SEGMENTS FREQUENCY_SEGMENTS
Diecast Classics Inc. 31 35.83870968 122138.14 868 Most_Recent High Occasional
 THANK YOU

MRAPROJECT-MILESTONE1-PUVYARAVI - SUPPORTING TABLEAU DOCUMENT