Shop News Jobs WorldWide Contact Us Search C-Byte Home Page Big Open Systems That Work
CompanyCustomersSolutionsAlliancesProductsServices
 
 
     
 
   Solutions/
 
      Application/  
      Decision/  
      Contact/
      e-business/  
      Data Center Ready/  
      Whitepapers/  
     
     
     
   
 

Customer Releationship Management  
 

Modeling Customer Relationships
(continued)

Elements of a Customer Relationship Management Database
There are a number of required features of a CRM database that the architect must integrate in order to support the marketing lifecycle. These are (in no particular order):

* Customer or prospect focus
* All relevant facets of the relationship over time
* Integration of external prospect lists
* Integration of external data classifications
* Integration of external data enrichment
* Ability to directly score the database and segment the database many times
* Ability to evaluate different campaigns and treatment strategies over time and across millions of transactions and customers
* Campaign management, prioritization, etc.
* Ability to predict future customer behavior based on past behavior

It is not possible to achieve all of the above features using either a flat file approach or a standard data warehousing approach alone.

Back to Top

Taking a Lifetime View of the Customer
In order to fully realize a CRM strategy, the marketer must have information that enables him to take a lifetime view of the relationship. A relationship is most usefully defined as the starting point at which the organization has an initial interaction with a prospect. This relationship then needs to be tracked as the prospect is encouraged to climb the loyalty ladder from prospect to customer and eventually to highly valued customer. The marketer needs to see and understand past events, contacts and purchase information in order to assess the current and future profitability of the relationship. The commonly used marketing analysis of recency, frequency and monetary value of transactions indicates some of the facets of the relationship that should be tracked.

In C-Byte's experience of facilitating client workshops to establish the business requirements for a CRM solution, four relationship facets appear common to most organizations. These facets are:

Product Holding-What products has a customer purchased and what products do they currently hold ?

Product Usage-How has the customer used that product? For example, can an increase in credit card usage be attributed to some prior interaction with the customer or some promotional activity?

Contacts-What has the organization's interaction with the customer been over time and what were the outcomes?

Events-What other events have occurred, either within the life of the customer (e.g., marriage) or externally to the relationship (e.g., competitor activity)?

Each of these facets may be treated by the modeler as an island of analysis linked centrally to an individual customer at a given point in time. For example, the marketer may take a point-in-time view of the relationship, a view over time or make prescient predictions for the future. Information about these four facets of a customer relationship enable the marketer to answer questions such as: How many customers have bought product X? How many customers display a repeatable purchasing pattern? How often have I contacted this customer and when? Who are my most profitable customers? What events or contacts occurred prior to customer defection?

The approach taken by C-Byte to support this kind of questioning is to place a customer table at the center of the model and to surround it with satellite dimensional schema (star schema) representing each facet of the relationship to be modeled. Modeling the facets of the relationship dimensionally allows who, what, when, where style analysis. For example: Which segment bought which products and what contacts preceded which purchase? Where do the contacts live, and how do they like to be addressed?

The customer-centric nature of the model also lends itself well to the prudent de-normalization of often-used facts, such as disposable income estimates, onto the customer table and helps facilitate the efficient extraction of contact lists and integration with statistical modeling tools, such as SAS or Unica. The customer-centric model also supports very well the iterative nature of the marketer's questioning, such as: How many customers hold product Y? Which of those customers are profitable? Which of those customers did I contact last week and which of them complained about the contact? It is also possible to assess what behavioral changes are exhibited as a result of identifiable interactions with the customer. Once the marketer has exhausted his questioning, which helps refine the contact list names, addresses and salutations may be simply extracted from the customer table using the relevant keys. Current suppression indicators and propensity scores may also be stored against the central customer record, allowing the possible automation of standard hygiene filtering.

The Customer-Centric Model at an Insurance Company
To see how this model might work, take the example of an insurance business. The firm's relationship with Mr. Jones begins when he makes an initial inquiry about health insurance via the organization's call center. This initial inquiry is the result of a press advertising campaign that reached Mr. Jones; this fact is recorded.

In response to his interest in the company's health insurance offering, the insurance business sends Mr. Jones an information pack. This step is also captured and recorded in the database. At this point, Mr. Jones does not have a product holding, but his name and address and contact records exist within the database. Mr. Jones does not respond to the receipt of the information pack, and after three months the marketer plans a campaign targeted at Mr. Jones and all the other Mr. Joneses who have interacted with the organization but not purchased any products in the last three months.

In this case, a query can be run against the database asking, Who has contacted us in the last three months with a contact type of inquiry? This query will generate a list of keys into the customer or prospect table, which, without further refinement, could be used to generate a contact list. However, it is more likely that the marketer's questioning will continue further—How many of these customers or prospects were sent an information pack? The result set from this query will be matched against the result set from the last query to further refine the list of keys. This process may be further refined by asking, How many people in this list do not have a product holding? Once the marketer has completed his refinement of the list, it is a simple, and highly performant, exercise to take the resulting list of keys and extract the name, address, salutation data, etc. from the central customer or prospect table and perform further filtering based on suppressions on the customer table or assigning customers to campaign cells for different treatments based on segmentation keys on the customer record. Once the contact list is finalized, the customer keys are used to populate the contact table and to record the fact of the outbound contact. By storing all of this data in a centralized relational database management system (RDBMS), it is a relatively simple matter to make this data available to sophisticated campaign management tools and statistical modeling tools. These tools interface easily with an open RDBMS, such as Oracle, and almost without exception, such tools feature native connectivity options.

Those readers familiar with the processing dynamics of most RDBMS will immediately spot a major dependency of this model—the various software components deployed to support the marketing lifecycle must allow the generation of interim result sets. This is absolutely crucial in order to support the marketer's analytical processes as he constantly shrinks and expands potential target lists, possibly to generate the required list size to match a budgetallocation. Already, a number of tools vendors are acutely in-tune with the mindset and thought processes of the modern marketer.

Back to Top

Previous | Table of Contents | Next

 

 
 
 
   
   
C-Byte Company, Inc.
 
  Good Morning it is 12:01 am in Las Vegas, Nevada, United States.    E-mail This Page  |  Digg This  |  Facebook  |  Twitter  |  Delicious

Ask Us a Question - Call Inside US 1-800-393-5804, Outside US 1-403-770-7818, or Contact Us
Copyright © 1989 - 2017 C-Byte Company, Inc.  ::  C-Byte Privacy Statement   ::   P.R.O. Program ::  RSS 1.0 Feed
This site is protected by copyright and trade mark laws under U.S. and International law. All rights reserved.
SecurityGroup Audits Systems at C-Byte Tackling...