CUSTOMER RELATIONSHIP MANAGEMENT: CONCEPTS AND TOOLS

1
CUSTOMER RELATIONSHIP MANAGEMENTCONCEPTS AND
TOOLS

• Chapter 5
• Customer Intimacy

2
Why is customer intimacy important?

• Customer data is needed for
• Operational purposes
• to deliver better experience to customers at the
  sales, marketing and service interfaces
• Analytical purposes
• to make sense out of customer behavior
  classifying, clustering, predicting
• Management purposes
• To help construct the overall CRM strategy
  customers, propositions, channels

3
Database structure

• Files (tables) hold information on a single topic
  such as customers, products and transactions
• Each file (table) contains a number of records
  (rows).
• In the customer database, each record (row) is a
  unique customer.
• Each record (row) contains a number of elements
  of data
• E.g. customers name, address, gender,
  date-of-birth and telephone number.
• These elements are arranged in common set of
  fields (columns) across the table.
• A modern customer database therefore resembles a
  spreadsheet.

4
7 steps to building a customer database
1. Define the database functions
2. Define the information requirements
3. Identify the information sources
4. Select the database technology and hardware
platform
5. Build or buy applications to access and
process information
6. Populate the database
7. Maintain the database
5
Database functions

• Operational
• A telecoms customer service representative needs
  to access a customer record when she receives a
  telephone query
• A hotel receptionist needs access to a guests
  history so that she can reserve the preferred
  type of room smoking or non-smoking, standard
  or de-luxe.
• A sales rep needs to check a customers payment
  history to find out whether the account has
  reached the maximum credit limit

• Analytical
• The telecoms company wants to know which
  customers are signalling an intention to switch
  to a different supplier
• The hotel company wants to promote a weekend
  break to customers who have indicated their
  complete delight in previous customer
  satisfaction surveys
• The sales rep wants to compute his customers
  profitability, given the level of service that is
  being provided

6
How customer data are often stored

• OLTP
• Operational data resides in an OLTP (online
  transaction processing) database

• OLAP
• Analytical data resides in an OLAP (online
  analytical processing) database.
• Information in the OLAP database is normally a
  summarised extract of the OLTP database, enough
  to perform the analytical tasks.
• The OLAP database might also draw in data from
  other internal sources, such as billing data.

7
Defining the information needed

• The information needed depends on
• The operational processes to be performed
• Sales, marketing, service
• The analytical decisions to be made
• Propensity to buy, potential to churn, credit
  risk
• Distinguish between need-to-know and
  like-to-know information

8
Common customer information fields

• Contact data
• Contact history
• Transactional history
• Intentions
• Needs
• Benefits
• Expectations
• Preferences
• Benchmarks

9
To understand needs, understand motivations

• Because of motivations are linked to some
  prefigurative force.
• motivation to buy or consume is driven by some
  pre-existing condition.
• a company buys spare parts for its equipment
  because of a history of down-time in operations.

• In order to motivations have a future
  perspective.
• motivation to buy or consume is driven by the
  desire to achieve some future condition
• a private individual might buy a second home in
  order to enjoy the tranquillity of its rural
  location.

10
Benefits vary across segments

• Customers buy products to experience the benefits
  they create.
• Customer A buys consistent product quality, which
  enables them to run their manufacturing processes
  with fewer disruptions
• Customer B buys the same product because of its
  variety of applications, thereby eliminating the
  requirement to maintain and manage complex
  inventory.

11
Millers expectations taxonomy

• The ideal level. What can be
• The predicted level. What will be
• The minimum tolerable level. What must be
• The deserved level. What should be

12
Olivers customer expectations hierarchy
What the customer wants
Ideal Excellent Desired Deserved Needed Adequate M
inimum tolerable Intolerable
Tolerance zone
Zone of indifference
What the customer predicts
13
Two expectations zones

• The zone of tolerance
• this ranges from what must be (minimum
  tolerable) to what can be (desired level).

• The zone of indifference
• this ranges around the customers judgement of
  what is a reasonable expectation of the supplier

14
Why are expectations important?

• Matching offers to the expectations, whether
• Ideal, desired, deserved, adequate
• Expectations change over time
• Ideal expectations decay into normality
• Not all attributes are subject to customer
  expectations.
• Customers usually have expectations of a number
  of attributes.
• Not all of these attributes are equally
  important.
• Expectations act as the basis for satisfaction
  judgements.
• Suppliers need to understand the limits to each
  customers tolerance zone and zone of
  indifference.

15
Preferences for.

• Communication medium?
• mail, telephone, email, etc?
• If email, is plain text or html preferred?
• Salutation?
• Miss, Ms, Mrs, first name, family name?
• Contact time and location?
• phone anytime for urgent product recall?
• mail to work for invoicing?
• face-to-face at branch for news about new
  products?

16
Desirable data attributes STARTS

• Shareable
• Transportable
• Accurate
• Relevant
• Timely
• Secure
• The attributes are enabled by the architecture of
  the CRM system

17
Identify information sources

• Internal data
• sales, marketing, service, finance data
• External data
• compiled data
• census data
• modelled data
• Secondary and primary data

18
Compiled list data for a dancewear company

• memberships of dance schools
• student enrolments on dance courses at school and
  college
• recent purchasers of dance equipment
• life-style questionnaire respondents who cite
  dance as an interest
• subscribers to dance magazines
• purchasers of tickets for dance and musical
  theatre

19
USA geo-demographic census data

• median income
• average household size
• average home value
• average monthly mortgage
• percentage ethnic breakdown
• marital status
• percentage college educated

20
Individual-level data

• Individual-level data are better predictors of
  behavior than geo-demographic data
• in the absence of individual-level data census
  data may be the only option for enhancing
  internal data
• can use census data about median income and
  average household size to predict who might be
  prospects for a car resellers promotion.

21
Modelled data PRIZM analysis of TW9 1UU, UK

• young professionals
• rented accommodation
• above average car ownership
• take foreign holidays
• read the quality press
• assigned to PRIZM code A101

• Lifestyle A (A-D)
• Income quintile 1 (1-5)
• Cluster type 1 (1-72)
• 0.34 of GB households
• Income rank 5 (1-72)
• Age rank 28 (1-72)

22
Secondary and primary data

• Secondary
• Secondary data are data that have already been
  collected, perhaps for a purpose that is very
  different from the CRM requirement.

• Primary
• Primary data are data that are collected for the
  first time, either for CRM or other purposes.

23
Primary data collection schemes for CRM programs

• Competition entries.
• Customers supply personal data on the entry
  forms.
• Subscriptions.
• Customers subscribe to a newsletter or magazine,
  again surrendering personal details
• Registrations.
• Customers register their purchase. This may be so
  that they can be advised on product updates
• Loyalty programs.
• New members compete application forms, providing
  personal, demographic and even lifestyle
  information

24
Database technology and hardware platform

• Relational databases are the standard
  architecture for CRM databases.
• Relational databases store data in 2-dimensional
  tables comprised of rows and columns.
• In a customer database, each row is a unique
  customer and each column contains some attribute
  of that customer.
• Each customer is given a unique identifying
  number.

25
Customer unique identification number

• Allows linkages to be made between several
  customer-related databases (e.g. transactional,
  product and service databases)
• Customer records can be linked in 3 ways
• One-to-one. Each record in one database can be
  linked to one other record in another database.
• One-to-many. Each record in one database can be
  linked to many records in another database
• Many-to-many. Each record in one database can be
  linked to many records in another database, and
  each record in that database can, in turn, be
  linked to many records in the first.

26
Criteria influencing choice of hardware platform

• Size of the the database.
• Even standard desktop PCs are capable of storing
  huge amounts of customer data.
• Existing technology.
• Most companies will already have technology that
  lends itself to database applications.
• Number and location of users.
• Many applications are quite simple, but the
  hardware might need to enable a geographically
  dispersed, multi-lingual, user group to access
  data for both analytical and operational
  purposes.

27
CRM applications 1

• Marketing applications
• market and customer segmentation
• campaign management
• direct marketing
• event-based marketing
• multi-channel marketing

• Sales applications
• managing the sales pipeline
• lead management
• opportunity management
• contact management

28
CRM applications 2

• sales management applications
• salesperson performance management
• workload allocation
• salesperson appraisal

• service applications
• contact centre management
• customer communications
• enquiry handling
• helpdesk management
• complaints management

29
Selecting the correct analytical applications

• how many variables need to be analysed at the
  same time?
• Univariate, bi-variate, multi-variate
• do you want to describe a set of data or to draw
  inferences about a population?
• what types of data are you analysing?
• Nominal, ordinal, interval, ratio

30
Populating the database

• Four methods for creating appropriately
  accurate customer records
• verify the data
• Double-keying
• validate the data
• Range validation
• Check for values that are missing (empty cells)
• Check against external sources.
• de-duplicate the data
• merge and purge the data

31
Maintaining the database

• Enter data from all new transactions, campaigns
  and communications immediately
• Regularly de-duplicate the database
• Audit a subset of the files every year
• Purge customers who have been inactive for a
  certain period of time
• Drip-feed the database
• Get customers to update their own records
• Remove customers records on request
• Insert decoy records, if the database is managed
  by an external agency

32
Single view of the customer
Retail store
Integrated customer database
CRM Strategy development and implementation
Party plan
Analytical and operational applications
Catalogue store
Web-site
Home shopping
External data
33
Data warehouses and data marts

• A data warehouse is a repository of large amounts
  of operational, historical and customer data.
• Data volume can reach terabyte levels, i.e. 240
  bytes of data.
• Attached to the front-end of the warehouse is a
  set of analytical procedures
• Retailers, home shopping companies and banks have
  been early adopters of data warehouses.
• A data mart is a scaled down version of the data
  warehouse.
• Data mart project costs are lower because the
  volume of data stored is reduced, and the number
  of users is capped
• Technology requirements are less demanding.

34
Data transformation before warehousing

• Data standardisation
• Personal data m/f, M/F, male/female
• Units of measurement metric/imperial
• Field names sales value, Sale, val
• Dates mm/dd/yy, dd/mm/yyyy, yyyy-mm-dd
• Data cleaning
• De-duplication
• Updating and purging
• Identify misuse of data entry fields e.g. use of
  phone field to record email address

35
Mining warehoused data

• Mining warehoused data can find
• Associations
• Sequential patterns
• Mining warehoused data can establish
• Classifications
• Clusters
• Mining warehoused data can enable predictions to
  be made

36
SEMMA - the SAS data-mining model

• Sample Extract a portion of the dataset for
  data mining
• Explore Search for trends and
  relationships
• Modify Create, select, transform variables
  with the intention of building a model
• Model Specify relationships between
  variables to predict a specific outcome
• Assess Evaluate the model

37
Responses to privacy concerns

• Self-regulation by companies and associations
• companies may publish their privacy policies and
  make a commercial virtue out of their
  transparency
• professional bodies in fields such as direct
  marketing, advertising and market research have
  adopted codes of practice
• Legislation

38
Scope of the OECD privacy principles, 1980

• Purpose specification
• Data collection processes
• Limited application
• Data quality
• Use limitation
• Opt-out, opt-in
• Openness
• Access
• Data security
• Accountability

39
Legislation guarantee these rights to EU citizens

• Notification
• Individuals are to be advised with without delay
  about what information is being collected, and
  the origins of that data, if not from the
  individual
• Explanation
• Of the logic behind the results of automated
  decisions based on customer data. For example,
  why a credit application was rejected.
• Correction/deleting/blocking
• Of data that does not comply with legislation.
• Objection
• Individuals can object to the way their data is
  processed (opt-out). Where the objection is
  justified, the data controller must no longer
  process the information

40
Obligations on data controllers

• Only collect and process data for legitimate and
  explicit purposes
• Only collect personal data when individual
  consent has been granted, or is required to enter
  into or fulfil a contract, or is required by law
• Ensure the data is accurate and up-to-date
• At the point of data collection, to advise the
  individual of the identity of the collector, the
  reason for data collection, the recipients of the
  data, and the individuals rights in respect of
  data access, correction and deletion
• Ensure that the data is kept secure and safe from
  unauthorized access and disclosure.

41
W3Cs approach to internet privacy (P3P) contains
3 elements (1)

• A personal profile
• Each Internet user creates a file consisting of
  personal data and privacy rules for use of that
  data.
• Personal data might include demographic,
  life-style, preference and click-stream data.
• Privacy rules are the rules that the user
  prescribes for use of the data, e.g. opt-in or
  opt-out rules, and disclosure to third parties.
• The profile is stored in encrypted form on the
  users hard drive, can be updated at any time by
  the users, and is administered by the users Web
  browser.

42
W3Cs approach to internet privacy (P3P) contains
3 elements (2)

• A profile of web-site privacy practices.
• Each Web-site discloses what information has been
  accessed from the users personal profile and how
  it has been used.
• Automated protocols for accessing and using the
  users data.
• This allows either the user or the users agent
  (perhaps the Web browser) automatically to ensure
  that the personal profile and the privacy rules
  are observed.
• If compliance is assured, then users can enter
  Web-sites and transact without problem.