Building a Web-Enabled Data Warehouse

Data Warehouse Development Approach

Advanced DataTools’ unique data warehouse development approach is based on years of successful work with many types of teams. Our approach ensures that:

• Results are quickly demonstrated in Phase 1 to facilitate refinement of the project plan and keep motivation high throughout the project.
• All perspectives on the project are explicitly included, from executives’ requirements, to the end-users’ needs and computing environment, and the data processing staff’s knowledge of the systems environment and its challenges.

ADTC’s six phase approach maintains focus on the critical success factors along the development path: committed user and technical staff involvement from the beginning, clear definition of scope to prevent paralyzing scope-creep, early executive review and buy-in to ensure priorities are met, careful attention to configuring a platform that will enable rapid response time to queries, intense scrutiny of the data loading and cleansing process to ensure data integrity from source to data warehouse, and documentation and training of technical and production staff and end users to guarantee active use, refinement, and custodianship of the data warehouse.

Data Warehouse Development in 6 Phases

Phase 1 is the Key to Success and User Acceptance

The objective of Phase 1 is to build a prototype data warehouse in eight weeks. The prototype has been instrumental in helping our customers deliver initial results to users quickly, and provides management and users with an early preview of the data warehouse. It serves as a foundation for the full production data warehouse system and allows customers to test the warehouse before full-fledged development is underway, rather than waiting 6 months to a year into the project only to find that the warehouse does not meet their requirements. The prototype consists of a one-time data download and does not include nightly updates.

For the prototype to accomplish its purpose, it is important that the scope of the data be limited and well defined. ADTC recommends that the customer select a subset of the planned production data warehouse for the source data. Information gathered during implementation of the prototype will then be used by the Project Team to anticipate the effort and next steps required to expand the prototype into a fully deployed data warehouse. Issues that arise during the prototype implementation will be documented and brought to the attention of the Executive Team.

Phase 1: Building the Prototype

Data Warehouse Vision and Design Workshops

The design of the data warehouse begins with a Vision Workshop that is fundamental to the success of the project. Instead of plodding through months of user interviews to gather project requirements, ADTC brings the most active data warehouse users and technical staff into the same room for an in-depth requirements workshop. The interaction between users from across the organization, technical staff, project management, and the ADTC team develops a broader, deeper understanding of the business and technical requirements, ensures the best possible design for the data warehouse, and develops ownership of the data warehouse at all levels of the organization.

Following the vision workshop, the technical members of the Data Warehouse Team tackle the systems design aspects of the project with a Design Workshop. They will build the data warehouse model, prepare for data transfer, lay the groundwork for software and hardware requirements, and address any anticipated barriers to the active and ongoing use of the data warehouse. This work goes through further refinement following prototype testing. These multi-day workshops with end users and technical staff are essential to a successful data warehouse project. Full participation in these workshops is required of all persons identified on the Project Team outlined below, including the end users who will be most active, the Customer Project Manager, and the Customer Technical Lead.

Team Effort

ADTC’s approach requires the full-time involvement of a Project Team, including end users of the data warehouse, technical staff from the Customer’s Data Warehouse Team, and ADTC consultants. It cannot be over-emphasized that one of the primary keys to success is the committed participation of designated end users from a wide variety of levels across the organization in the Vision Workshop.

ADTC staff will perform three roles:

• Data Warehouse Architect
• Project Leader and Brio Consultant
• Senior Systems Analyst and Database Administrator

Customer staff roles are crucial to the success of the project:

• Executives
• Data Warehouse Project Manager
• Data Warehouse Technical Lead
• End Users
• DBAs, Systems Staff, and Programmers

Data Warehouse Architecture

The hardware and software architecture for a typical data warehouse project encompasses a staging area for loading and checking the data, and a production database for the data warehouse. To allow end users to query the database from a web browser, a web server and a Brio Enterprise server are required.

• Staging Database
• Production Data Warehouse Database
• Web Server
• Brio Enterprise Server