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/  
     
     
     
   
 
 
C-Byte Delivers High-Availability

Customer Releationship Management  
 

Modeling Customer Relationships
(continued)

Integrated Infrastructure Supports Marketing Process
In the past, database marketing solutions often focused on individual user communities participating in the overall marketing process. While this focus has managed to hit the sweet spots of these often isolated communities (e.g., marketing analysis or campaign management) and to temporarily satisfy parochial needs, it has left a troublesome legacy for the integrator of the technology layer who seeks to accelerate the marketing cycle, empower the marketer and reduce the marketing department's dependency on highly skilled and expensive (and often obstructive) database experts. Such function-focused solutions have ensured that the walls that block the implementation of a virtuous circle of continuous improvement in the marketing process remain solid. The proliferation of file formats, APIs and unnecessary processing layers needed to integrate these elements have delivered a full employment charter for those who wrangle with the complexity of the technology layer at the expense of marketing responsiveness and creativity.

C-Byte's solution to such technical anarchy is to focus firmly on a technological infrastructure that supports and integrates the overall marketing process, and underpins the progressive development of a relationship management strategy. The use of a centralized relational database and open systems to manage customer data, contact history and relationship history allows the easy integration, at the data level, of the various technologies deployed at different stages in the marketing process. Analysts' models may be stored alongside the actual data, and scoring and segmentation keys can be made directly available to campaign management and campaign scheduling software. The automation of routine communications is simplified and database triggers can be utilized to make marketing more event driven.

Back to Top

Typical Data Modeling Challenges
This section details some of the data modeling challenges, which, in C-Byte's experience, are common across a number of industries and organizations.

Householding
The grouping of individuals by household or relationship patterns is often a difficult process in product-focused legacy systems. These systems often have great difficulty in even identifying the individual responsible for purchasing a given product. The benefits of groupings for the relationship marketer are many:

* Avoidance of unnecessary duplicate contacts per household
* Understanding loyalty patterns among relationship groups
* Identification of cross-sell and up-sell opportunities (e.g., family policies, etc.)
* Identification of significant life events (coming of age, birth, marriage, etc.)
* Analysis of geodemographic data by household

Multiple households can be problematic for both the marketer and the system designer. Individual customers may have multiple addresses, each of which is related to the customer via the product holding. For example, Mr. Jones has a main residence in the city and a weekend retreat by the coast. Mr. Jones has a household insurance policy for each address. An insurance marketer may wish to sell Mr. Jones a life insurance policy. However, for the modeler, a household is just a simple grouping of individuals. Specific business questions must be answered in order to track the household movements of individuals. The difficulty is in the actual identification of a household—particularly in high-density urban residential areas or areas with a highly transient population.

There are several approaches to handling customer householding, de-duping and geocoding challenges. These include:

* Service Bureau operations
* Integrating specialized software tools to perform this function on a regular basis (this also requires process integration for proper and effective handling)

A number of marketing data processing bureau services perform household identification, based on, for example, electoral register information, etc. However, such matching is never 100 percent accurate.

Products Held by Groups of People
Certain types of products, for example joint bank accounts, introduce a many-to-many relationship between product holdings and persons. This fact, if modeled literally, can cause performance problems in the database and confuse campaign management and extraction tools seeking to identify a single prospect. This is particularly true in cases where organizations are transitioning to a customer-focused marketing strategy yet still require the ability to market in the interim period based on product holding attributes. This situation is common in large businesses that cannot possibly switch from a product to a customer focus overnight. The only answer to this problem is a business one. Identifying a primary marketing contact for a product holding can simplify the problem in some cases.

Person Matching
Another key challenge for the designer of a CRM database is the identification of individuals. Often, seemingly multiple individuals on the database are in fact the same person, albeit at a different point-in-time, or with a different product holding, or at a different address. Organizations with multiple operational systems serving multiple customer touch points often find that the non-uniformity of input validation across these systems leads to situations where Mr. John Jones, Mr. J. Jones and Mr. J. B. Jones at the same address could perhaps be one, two or three actual people. This problem is further exacerbated when external prospect lists are brought into the database. Once again, the modeler can incorporate a simple grouping of people within the database design but the problem is identifying the actual grouping. In some cases, C-Byte has allowed a "degree of confidence" value to be assigned to the grouping record to provide the marketer with a coefficient that validates assumptions. The business rules for deriving this coefficient clearly evolve over time, and can result in the creation of specific profiling questions targeted to specific customers during interactions.

As with householding, some marketing data providers can perform unique person identification based on postal lists, real estate listings, electoral rolls, and other data. This identification activity can be cumbersome as it involves exporting and re-importing data periodically. If the grouping of seemingly multiple individuals into one is handled as a grouping table,the impact on, for example, referential integrity within the database can be minimized. However, this kind of group can also make the model more complex—with a possible impact on performance.

Unfortunately, there are no magic cures for the problem of person matching, and the database modeler should be wary of the purveyors of such cures.

Classing and Banding
A number of marketing database designs use fields such as "date of birth" or "age" on the customer record. Though there is a clear use for such fields, marketers rarely wish to contact people who are, for example, 51 or 23 years of age. Usually, the marketer wants to target people aged between 25 and 35 or those who are past retirement age. Such targeting calls for some sort of banding of customers to reduce wasted processing and simplify the process for the marketer.

Age is not the only candidate attribute for banding. The modeler should seek to understand other candidates and include these in the model.

Regularly Used Measures
Initially, and over time, the modeler of the customer database should seek to identify those frequently asked marketing questions, such as: Who earns more than $20,000? Who has made more than four insurance claims in the last period, etc.? It makes little sense to have multiple marketing campaign designers all scanning the product usage table over and over again. This can be avoided by denormalizing regularly used measures directly onto the customer or prospect record.

Suppressions
Most organizations are able to identify a number of standard reasons for suppressing marketing communications. Suppressions can range from blanket "do not communicate at all" indicators to "do not market a specific product" to this individual. These suppressions should be held directly on the customer or prospect record to enable swift and easy filtering of targets.

Back to Top

Summary
While both flat file and standard data warehousing approaches to the customer database will allow analysis of customers and the selection of target lists, neither approach will, on its own, support the management of customer relationships over time. Likewise, neither will integrate all components of the marketing process in the most efficient way.

The template presented in this paper may form the basis of the data architect or analyst's initial attempts to define data structures, which will support both of the above objectives. This template reflects the work C-Byte has done with a number of major organizations to support their database marketing activities and to drive the strategic implementation of Customer Relationship Management at both the business and the systems levels.

CRM is an emerging strategy and as such requires a fresh approach to systems design, along with the flexibility to accommodate unexpected change. Many piecemeal or point solutions in the market fail to take an integrated view of the entire marketing lifecycle and focus only on data structures to support their own specific components of that lifecycle. As CRM matures as an operational reality, it is imperative that organizations have an integrated view of business processes and data. Failure to take an integrated view of requirements will lead to significant effort and cost reengineering the organization's marketing databases—sometimes comprising many terabytes of data.

Back to Top

Previous | Table of Contents

CP-1340 12/01

 

 
 
 
   
   
C-Byte Company, Inc.
 
  Good Evening it is 11:42 pm 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...