A successful BI/DW project depends on the ability and commitment to systematically plan for and assess all of the elements within the organization that could influence the effectiveness of its outcome. Whether through budget constraints, time pressure, inadequate experience and expertise, or the inability to obtain the needed data and resources, a project will fail if any critical element is ignored or not given adequate consideration. Requirements should be set up which are measurable and achievable.
After defining all of the phases, activities, and deliverables, priorities must be set. Timelines and maturity models should be established and agreed to by all critical players, complete with key milestones and intermediate checkpoints. Resources should be scheduled and available at the right times to meet each of the objectives, with early warnings to indicate if the project start date is moved.
Of course, a full commitment throughout the organization is critical, from senior management down to the end user, and all parts in between. A tested and proven methodology executed by people with the experience and expertise to carry it out, and a complete meeting of the minds on project scope are of vital importance. The core team must also understand the business processes and the data that they are gathering, and be aware of any unique environmental aspects and any interdependencies or interconnections that may exist. The plan should be broken into manageable phases, such as business processes, data sources, user communities, data complexity, etc.
The Future of Big Data
With some guidance, you can craft a data platform that is right for your organization’s needs and gets the most return from your data capital.
An essential component of the plan is the establishment of initial training of users that covers the tools, techniques, methodology and the data itself. (It should not be assumed that the users have a full understanding of the data, because they frequently do not.) Build on that early training for smoother progress and as a foundation for gradually increasing users’ knowledge, understanding, and comfort level. Earlier phases of the project should be of lesser complexity to allow them to gain a further understanding of the environment before encountering the tougher challenges ahead. Ensure that the plan is adjusted based on the tools and data that are available for use.
There should be a strong focus on the quality of the data itself, and a plan should be in place to filter out unwanted and erroneous data, and to ensure that it conforms to preset standards. The solution design must have sufficient flexibility to support a dynamic environment, and to provide scalability of the data model and architecture.
A rigorous, practical and efficient testing methodology should be executed at logical points along the way to verify the expected outcomes, and to discover any anomalies as early as possible in the process.
A change management process should be established that clearly delineates the support resources for dealing with daily activities as well as conflicting dependencies and urgent matters that may arise. Ensure a chain of communication that can keep appropriate parties informed on a daily basis, and can also quickly escalate critical issues as needed. Allow for substitution of personnel as needed, and for cross-training of skills, so that the absence of one party will not bring the project to a halt.
Throughout the project, repeatable processes should be implemented, which will make it easier to ensure that all parties are on the same page, will simplify documentation, and allow the process to be run again at any time. It will also make for more effective updates and enhancements in the future.
