CRM and Information Visualization Gürdal Ertek, Ph.D. Tuğçe Gizem Martağan Tuğçe Gizem Martağan
1
Customer Relationship Management (CRM) Traditional Marketing
CRM
Goal: Expand customer base, Goal: Establish a profitable, increase market share by longterm, onetoone mass marketing relationship with customers; understanding their needs, preferences, expectations Product oriented view
Customer oriented view
Mass marketing / mass production
Mass customization, onetoone marketing
Standardization of customer needs
Customersupplier relationship
Transactional relationship
Relational approach
2
What is CRM? “The approach of identifying, establishing, maintaining, and enhancing lasting relationships with customers.”
“The formation of bonds between a company and its customers.”
3
Strategies in CRM for Mass Customization • • • •
Prospecting (of firsttime consumers) Loyalty Crossselling / Upselling Win back or Save
4
5
6
The Marketing Perspective CAMPAIGN MANAGEMENT RECENCY FREQUENCY MONETARY VALUE METHOD CUSTOMER VALUE METRICS
7
Campaign Management: The Marketing Perspective • • • •
Developing effective campaigns Effectively predicting the future Retaining existing customers Acquiring new customers
8
Campaign Management: The Cap Gemini Model KNOW
TARGET
Understand market and consumers’ needs and preferences
( Offer is developed ) Define market strategies
Exploit customer intelligence, Use channel integration Perform segmentation
SERVICE
SELL
Retain customers by:
Acquire customers
Loyalty programs Communication Service forces
Use sales force effectively Develop marketing programs 9
Campaign Management: The Marketing Perspective The marketing manager... 1. Defines objectives 2. Identifies customers 3. Defines communication strategies 4. Designs/improves products/offers/services/promotions 5. Tests the impacts of her decisions 6. Revises her decisions for maximum effectiveness 10
Campaign Management Step 1: Define Objectives Targeting Existing Customers Retention Strategy
Targeting New Customers Acquisition Strategy
Creating Loyalty? Increasing the satisfaction level? Crossselling or Upselling?
Target customers that show characterstics similar to existing groups of customers
11
Campaign Management Step 2: Identify Customers Perform SEGMENTATION • Define the right customers • Use information of past transactions as key for making predicting future ones • Define the segments and their characteristics • Develop customized marketing strategies for the different segments
12
Campaign Management Step 3: Communication Strategies • Which message should be transmitted? • Which channel should be used?
13
Campaign Management Step 4: Design the Products, Offers, Services and Promotions • Analyze the price, time period, risks, marketing costs • Define the product / offer / service / promotion and its general structure • Identify effective use of sales and communication channels
14
Campaign Management Step 5: Test the Impacts • Impacts of the decisions have to be tested and and assessed on a sample
15
Campaign Management Step 6: Revise the Decisions • Make revisions to the targeted offer / service / promotions • Finally apply the decisions to the whole segment or population
16
17
18
RFM Method (Recency, Frequency, Monetary Value ) • Recency – When was the last customer interaction? • Frequency – How frequent was the customer in its interactions with the business? • Monetary value of the interactions
19
RFM Method (Recency, Frequency, Monetary Value ) Marketing Problem: A firm has sent email to 30,000 of its existing customers, announcing a promotion of $100. 458 of them responded (1.52% of the customers) Is there any relation between the responding customers and their historical purchasing behaviours? 20
RFM Method: Recency Coding • 30,000 customers are sorted in descending order with respect to their most recent purchases • Sorted data is divided into 5 equal groups, each of them containing 6,000 people • Recency codes are assigned: Top group has code 5, bottom group has code 1 21
RFM Method: Recency Coding Response %
Recency Results 4.00
• According to analysis based on customer recency, the group having the highest recency group has also the highest response rate
3.1
3.00 2
2.00
1.5
1.00
• Remark: (3.10% + 2.00% + 1.50% + 0.62% + 0.38) / 5= 1,52% which is the response rate 0.38
0.62
0.00 5
4
3
2
1
Recency code R
• Strict Rule: Ones who have purchased recently are much more willing to buy new products than others 22 purchasing in the past
RFM Method: Frequency Coding • Sort the 30,000 customers with respect to frequency metrics. – Frequency metrics: Average number of purchases made by customer in a time period t – Sort customers in descending order with respect to their purchase frequency.
• Assign them to 5 groups, top %20 in the first frequency group. • Assign frequency codes such that the top group has code 5 and the bottom group has code 1. 23
RFM Method: Frequency Coding Frequency Results 3
2.8
Response %
2.5
2.1
2 1.5
1.3
1
0.8
0.9
0.5 0 5
4
3
2
Frequency code F
1
• It is observed that highest response rate is from the customers having highest frequency • Frequent people respond better than less frequent ones but differences between groups are less than the ones in the recency • The lowest frequency group always contains new customers 24 • That is why it is named RFM
RFM Method: Monetary Value Coding • The same process as recency and frequency coding • Sorting is done with respect to monetary value metric – Monetary value metric is the average amount purchased in a time period t
• At the end of the monetary value coding, assign monetary value codes M = 1,...,5 to groups according to their groups. 25
RFM Method: Monetary Value Coding Frequency Results
2.5 Response %
2.1
2
1.8 1.4
1.5
1.2
1.1
1 0.5 0 5
4
3
2
1
• It is observed that highest response rate is from the customers having highest monetary value • Unlike the recency case, there are not big differences between groups
Monetary value code M 26
RFM Method: Putting the Codes Together • At the end of the monetary coding firm obtain R F M metrics for customers. Each customer belongs to one of 125 possible combinations of the RFM values: Database
1
2
3
4
R
5
21
22
23
24
231
232
233
234
25
235
F
M
27
RFM Method: STEPS • Create 3 digits RFM codes cells • All cells having the same number of customers in them • RFM values are used to define group of customers that marketing campaign should target or should avoid • Used for identifying customers having high probability to respond to campaigns: 555’s response rate > 552’s > 543’s >541.... • Increase the response rate • Increase profitability
28
Customer Value Metrics • Critical measures used to define customer worth in knowledgedriven and customer focused marketing
29
Customer Value Metrics: Size of Wallet J
• Size of wallet = å S j j = 1
S j = Sales to focal customer by firm j • Assumption: Firms prefer customers with large size of wallet in order to retain large revenues and profits
30
Customer Value Metrics: Individual Share of Wallet (SW) • A proportion expressed in terms of percentage, calculated among buyers • Measured at individual level • A measure of loyalty • Can be used in future predictions • Different from the “market share”, which also considers customers with no purchase • Individual share of wallet % =
S j J
å S
j
S j = Sales to focal customer by firm j
j = 1
31
Customer Value Metrics • Share of wallet and size of wallet should be analyzed together because... Size of Wallet Customer 1 $500
Share of Purchases Wallet 50% $250
Customer 2 $100
50%
$50
32
Customer Value Metrics: Transition Matrix • Shows expected share of wallet from multiple brands • Depicts consumer’s willingness to buy over time • Transition probability from B to A, than from A to C: 10%*20% = 2% Brand A Brand A 60% Brand B 10%
Brand B Brand C 30% 20% 80% 15%
Brand C 20%
15%
70%
33
The Engineering Perspective DATA MINING
34
Data Mining • Collection, storage, and analysis of –typically huge amounts of data • Data readily resides in the company’s data warehouse • Data cleaning is almost inevitable
35
Data Mining Goals of Data Mining • • • •
Developing deeper understanding of the data Discovering hidden patterns Coming up with actionable insights Identifying relations between variables, inputs and outputs • Predicting future patterns 36
Data Mining: Steps • • • • •
Data selection Data cleaning Sampling Dimensionality reduction Data mining methods
37
Data Mining: Methods • Exploratory Data Analysis • Segmentation – Cluster Analysis – Decision Trees
• • • •
Market Basket Analysis Association rules Information Visualization Prediction – Regression – Neural Network – Time Series Analysis
38
Information Visualization Data mining algorithms... • Can only detect certain types of patterns and insights • Are too complex for end users to understand
39
Information Visualization • A field of Computer Science which has evolved since the 1990s. • Before 1990s: Graphical methods for data analysis to pave the way for statistical methods • After 1990s: – Computer hardware has advanced with respect to memory, computational power, graphics calculations – Software has advanced with respect to user interfaces – Data collection systems have advanced (barcodes, RFID, ERP)
40
Information Visualization • The analyst does not have to understand complex algorithms. • Almost no training required. • There are no limits to the types of insights that can be discovered. 41
Case Studies Analysis of Supermarket Sales Data
42
The Data Field Name
Desciption
TRANSACTION_ID
Transaction ID
PRODUCT_NO
Product Number
43
Frequent Itemsets Frequent Itemsets
44
Frequent Itemsets Frequent Itemsets
45
Association Rules Association Rules
46
Case Studies Analysis of Spare Parts Sales Data
47
The Data Field Name
Desciption
DEPOT
Depot ID
SKU_NO
SKU (Stock Keeping Unit) Number
VENDOR
Vendor (Customer) Number
DAY
Day of the month (1,...,31)
MONTH
Month of the year (1,...,12)
YEAR
Year (ex: 2002)
QUANTITY
Quantity required
UNIT_PRICE
Price of one unit of product in YTL*
REVENUE
Revenue from the order line 48
Assumption: Each customer gives at most one order each day.
Determining Top Products: Pivot Table for Determining REVENUE_SUM
49
Determining Top Products: Pivot Table for Determining COUNT (Frequency)
50
Determining Top Products: Scatter Plot
51
Seasonality of Top Products
. . .
52
Seasonality of Top Customers: Pivot Table
53
Cumulative % Revenue
Determining Top Customers: Pareto Curve (ABC Analysis) 100 90 80 70 60 50 40 30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
Cumulative % Customers Revenue
54
Seasonality of Top Customers: Starfield Visualization
55
Case Studies Analysis of ÖSS 2004 Data
56
The Data Field Name
Desciption
HS_NAME
High School Name
HS_TYPE_TEXT High School Type UNIV_NAME
University Name
UNIV_DEPT
University Department
RANK_SAY
Rank According to Sayısal Score 57
Pareto Squares L Y (L)
H
s T
Y 5 (H)
58
Pareto Squares: Model Definitions
59
Pareto Squares: Optimization Model
60
General Insights
61
Benchmarking Highschools
62
Benchmarking Departments
63Â
Relationship Management
64
References • Berry, M. J. A., Linoff, G. S. (2004) Data Mining Techniques. Wiley Publishing. • Ertek, G. Visual Data Mining with Pareto Squares for Customer Relationship Management (CRM) (working paper, Sabancı University, Istanbul, Turkey) • Ertek, G., Demiriz, A. A framework for visualizing association mining results (accepted for LNCS) • Hughes, A. M. Quick profits with RFM analysis. http://www.dbmarketing.com/articles/Art149.htm • Kumar, V., Reinartz, W. J. (2006) Customer Relationship Management, A Databased Approach. John Wiley & Sons Inc. • Spence, R. (2001) Information Visualization. ACM Press. 65
66