Data Warehousing Architecture: Options
Enhancing Customer Relations Relations with Customer Data
Mar 30, 2009
This strategy envisions a system that not only collects and stores information, but also provides the tools to enable businesses to use this information to manage customer relationships that will benefit both customer and corporation. This enterprise-wide capability enables organizations to gain competitive advantage in their individual markets,through improvements in all aspects of customer relationship management.
Enterprise-wide System
The original concept of a data warehouse was a single, enterprise-wide system that stored all corporate data for access by users throughout the organization. However, the real world of data warehouse and data mart system implementation has evolved to a point where the technology is being applied in several different forms to meet different application requirements and varying corporate objectives.
The enterprise data warehouse (EDW) option is the data warehousing technology that provides the broadest and most effective CRM solution. The EDW is the repository for data from enterprise-wide sources, as well as from any other source of customer data, and in its final form represents the totality of up-to-date customer data -- demographic, transactional, and current, real-time information from all customer touchpoints.
Another definition of the EDW is that it is a “common and unique repository for enterprise information. It is a read-only environment made up of detailed and aggregated data that is fully cleansed and integrated; and includes extensive detailed history of transaction level data.” The advantages of EDW architecture are:
- Presents a single version of customer data
- Provides one set of extraction processes and business rules
- Uses common semantics
- Offers a centralized, controlled environment
- Enables easy creation and population of subset data marts
- Designed with a single metadata repository.
The disadvantages of EDW architecture are:
- Extremely resource intensive
- High cost to implement
- Requires enterprise scale systems and resources
- Risk of storing all corporate data in one system.
Special Function Data Warehouses
There are specialty data warehouses which indicate the range of data warehousing technology and applications that have evolved over the past several years. These include the following architectures:
- Federated Data Warehouse/Data Mart (FDW/FDM)
- Incremental Architected Data Mart (ADM)
Other specialized data warehouses that have evolved and been implemented in some organizations are do not meet the criteria required for full implementation of the CRM solution, by maximizing the use of customer data throughout the enterprise. These include: the data stage/data mart (DSDM); and the distributed data warehouse/distributed data mart (DDW/DDM) architectures.
Federated Data Warehouse/Federated Data Mart
In many organizations, multiple teams undertake data warehouse projects, resulting in multiple data warehouse systems across the enterprise. Although there can really be only one enterprise data warehouse, with all other entities being subsets or incremental data marts, few organizations have strictly defined their systems in these terms, so that many enterprises around the world will have two, six, or a dozen or more ‘data warehouse’ systems.
One of these specialty warehouses, the next evolution of the EDW architecture, is a federated system of data warehouses or data marts (FDW/FDM), a data warehousing solution that offers many of the features of the EDW. The FDW/FDM architecture is marked by the characteristics of sharing common data points among multiple data warehouse or data marts, eliminating redundancy and ensuring a consistent and unique version of complete customer data throughout the organization.
The Advantages of FDW/FDM architecture are that it provides
- Common semantics and business rules
- A single set of extraction processes
- Decentralized resources and control
- Parallel development process
The disadvantages of the FDW/FDM arhitecture are:
- Difficulty in coordinating efforts
- Political and ownership issues among departments
- Agreement required on architecture, business rules and semantics
- Complex technical environment
- Needs multiple metadata repositories
A significant factor in the selection and development of a data warehousing strategy for a corporation is the examination of these various architectures to determine which one best fits the corporate culture.
 |
