Continue in 2 seconds

Processes for Effective Execution of an Enterprise-Wide Business Intelligence and Data Warehousing Program

  • December 07 2006, 1:00am EST

My recent online column "Making the Case for an Enterprise-Wide Business Intelligence (BI) and Data Warehouse (DW) Capability" (June 2006) highlighted the need for businesses to "get serious" about their approach to developing an enterprise BI and DW capability. When pursuing this capability it is important to adopt a holistic view, followed by disciplined investment and execution. To develop the future vision for this capability, you should consider seven interrelated areas:

  1. Strategy
  2. People
  3. Process
  4. Metrics
  5. Applications
  6. Data
  7. Architecture

This column explores the key considerations of the "process" focus area.
Because enterprise BI and DW is a journey consisting of many projects over time, many processes should be implemented to ensure that the BI program runs like a well-oiled machine. These processes help the team be more efficient, effective and consistently deliver quality solutions.

Choosing a Course of Action

In my "people" column, I mentioned creating an advisory council for prioritization of activities of the BI team. This council should consist of high-level leaders from each of the major functional groups across the organization that have the authority to make decisions about the BI budget. Choose these people based on their influence within the organization and their passionate interest in using information for better decision-making. They should have opinions about the most important priorities for their function and the entire business because they will be expected to work together to come up with one list of priorities. The advisory council should meet quarterly as part of a regular process to confirm priorities and adjust as necessary.

Guiding How Things Get Accomplished

First, you will want to have a standard implementation methodology so projects become more predictable, especially for the people who are managing the projects, for those working on them and for the customers of the projects. One successful approach is to start BI projects with a prototyping activity. This would involve quickly creating a first pass data mart, possibly a cube and some reports. This prototype can be extremely valuable in helping the business customers to understand more about their reporting and analysis requirements. In addition, the prototype can help them get a feel for what the solution may be like, and it also helps the BI project team understand more about the data needed to meet these requirements as well as where the quality issues and gaps are in the existing data sources. Creating a prototype also serves as a valuable momentum builder to the business customers who start to see tangible progress early in the process. The lessons learned from this prototype can help drive the additional requirements gathering and the scope prioritization of the items to be completely built out and automated in the first full release.

The prototype would be followed by a more formal waterfall-style project methodology with requirements definition, design, development, testing and deployment activities. At the end of each of these phases, the project tasks, time lines and costs should be re-estimated and reviewed with the customers to make sure all parties are aware of what is going to happen next and what expectations to have. In situations where there are many stakeholders, it is a good idea to create some kind of "delivery agreement" document at the end of the design phase in addition to technical design documents. This would outline exactly what the business customers will get so that there are no surprises at the end.

Toward the end of the design and development phases, it is recommended to have technical checkpoints to ensure that the designs and "code" are following the standards that have been defined as part of the BI program. This will help to ensure that all projects are being implemented consistently. This technical consistency makes it much easier for a new BI solution to be transitioned to a support team which would be focused on keeping the new solution running smoothly in production and making minor enhancements and fixes. This technical consistency also makes it easier to transition developers into or out of the team, because each BI solution will look and feel like all the others under the covers.

If the IT applications teams have a standard methodology, it is also a good idea to have some tasks added early in their projects to review any requirements that seem to be reporting and analysis related. Upon review, the teams may jointly decide that these are a better fit for the BI team to handle. This can serve as an early warning system to prevent more shadow BI solutions from being implemented within the IT group.

Throughout BI and DW project execution, there should be a regular process for program management. This would involve all BI project managers, technical architects, support team leaders and other team leaders getting together to review status of all in-process activities. This is very helpful to ensure that project interdependencies are properly coordinated, to address common roadblocks affecting multiple projects, to coordinate staffing moves and to enable the project leaders to learn from each other so they can better anticipate risks on their own projects.

Ensuring Quality and Secure Information

Processes should be put in place for businesspeople to participate in the management of master data, defining data mapping rules between systems and managing data security. These activities are typically referred to as "data governance" processes in the industry parlance. The BI team can provide tools to help automate these processes and make them as "foolproof" as possible, but it is important that the execution of these processes is primarily owned by people from the business groups because the business needs to be accountable for the information contained within the BI and DW solutions.

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