Did you ever notice how much we use the word "MOST" in our day-to-day lives? MOST of the company employees are happy. I want the cookie with the MOST chocolate. The one with the MOST toys wins. I want the MOST return I can get from my investment. In each case, the word is linked to some sort of direct or implied success. It is a very important word; yet when taking on a data warehouse project, many companies don't look at how to get the MOST out of their data warehouse.

It is the responsibility of the data warehouse project manager to assure that an organization is in a position to get the MOST from its data warehouse project.

For the purposes of this article, MOST is an acronym that represents four different views of an organization:

M ­ Management
O ­ Organization
S ­ Social
T ­ Technology

MOST is critical to the sponsorship and enduring success of a data warehouse project. As outlined in this column, IT staff are proficient at addressing the data, application and technical architecture components. Often the support architecture component is either addressed poorly or neglected entirely.

During the data warehouse architecture review and design stage of a project, four distinct components of architecture should be considered. The outcome of the architecture review and design is identification of those components that will require upgrade or acquisition. Each of these architecture components must be reviewed against all four letters of MOST. The four stages that comprise every data warehouse architecture should include:

Data Architecture. Defines the sources and stores of business information. Includes quality and management standards for data and meta data.

Application Architecture. The software framework that guides the overall implementation of business functionality within the warehouse environment. Controls the movement of data from source to user, including the functions of data extraction, data cleansing, data transformation, data loading, data refresh and data access.

Technical Architecture. Provides underlying computing infrastructure that enables the data and application architectures. Components include platform/server, network, communications and connectivity, hardware/software/middleware and DBMS. The technical architecture must be sufficiently robust to address scalability, capacity, volume, performance, availability, stability, chargeback and security.

Support Architecture. Software components, policies and procedures for backup/recovery, disaster recovery, performance monitoring, data archiving and version control/configuration management. Organizational processes necessary to effectively manage the technology investment and those processes that are prerequisites of the project.

It is seldom that a data warehouse project is initiated without the knowledge and support of business and/or IT Management. It is difficult to imagine a project manager succeeding without considering the requirements of management during the support architecture analysis. Requirements of management include establishment of data quality programs and support of the usage of the data warehouse to develop business strategy. A crucial requirement of management is to champion the data warehouse project.

In a similar manner, Technology is always addressed in a project, since it is impossible to develop a data warehouse without technology. Often, consultants or staff are selected based on their knowledge of the selected technology.

So, both the Management and Technology components of MOST are normally addressed during the data, application and technical architectures, but what about the O and S? If you miss the O and S, you'll only get M**T from the project. The best way to assure the MOST from your data warehouse project is to perform quality focused analysis and definition of the support architecture.

The Organizational component of MOST addresses the organizational impact of the data warehouse. For example, a marketing data warehouse may reduce the amount of time for management to determine the effectiveness of a recent campaign. Increased knowledge could reduce the life span of a campaign, allowing additional campaigns to be conducted during a year. But, knowledge has a price. It could mean additional work for the marketing staff to prepare a new campaign or adjust a current one. The result of the business intelligence gained could actually cause an overall increase in the total number of hours required for the marketing projects, forcing the staff to work harder or recruit additional staff. Conversely, the data obtained from your new data warehouse project may alter or even eliminate the need for certain staff in the organization. Other issues that are considered during the review of the organizational component include the assignment and use of the project steering committee, staff recruitment and training and establishing the data warehouse help desk.

Finally, an analysis of the Social impact of the project on the organization should be considered. When looking at the social components, determine how the "feel" of the organization may change. How will the staff react to the organizational change resulting from the elimination of positions due to the warehouse? Does staff see the warehouse as an opportunity for career growth (as an analyst), or do they see it as an evil giant that may end their career (as a clerk)?

Social changes will also result in decisions based on new business intelligence versus the old use of hunches. Being able to perform analysis and quickly generate custom reports will result in the ability to rapidly change the business direction of the organization. Training of users is a key component to the social impact that the project will have on the organization. Training, or lack thereof, will determine how well the new project is embraced by users.

A simple four-letter word: MOST. But if you do not consider all four letters, your project will likely generate a different four-letter word: FAIL. If you do not consider all four components of your warehouse project (Management, Organizational, Social and Technology) against all four architectures (data, application, technical and support), you'll never get the MOST out of your data warehouse project.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access