Scope is any business intelligence (BI) project's biggest killer. This includes scope from a business requirements angle, as well as the scope of technical delivery.
This problem normally arises when a BI project is delivered according to high-level business terms, and not keeping up and coping with business and system changes. Other killers are scope-creep over fixed timelines or budgets, late notification of source system changes, and technical or data problems not foreseen during planning.
The size of the project must never be underestimated.
Ren'e Muiyser is principal BI consultant at PBT Group.
With this in mind, key factors to ensure the project's scope is managed from a technical perspective must be established during the initial phases of the project. In the same way that a project definition report (PDR) establishes the scope of the business requirements and the project management aspects - the project methodology, principles and strategy documents must establish the development and deployment scope.
A methodology needs to ensure an architecture framework is put in place that allows for the maximisation, re-use and the best-practice standards to be positioned and highlighted across the development efforts. Most of all, it needs to prove the functionality through comprehensive systems testing to establish a proven BI application process with optimal re-usability.
The application of the methodology should develop the capability to incrementally speed up the deployment processes. This needs to be done in a way that the solution conforms to architectural and technology standards, adheres to business requirements, the quality of all deliverables (both documentation and code), is of a high standard, and the solution entrenches optimal re-usability.
In alignment
Key principles for a BI project must be to deliver an integrated collection of technology, architecture, processes and support to allow the business to build and utilise a more effective solution across various business units.
A framework benefits from a set of solutions that work well and have been specifically designed to offer the enhanced BI capabilities required for the business. This does not mean there is not a level of flexibility to support the various unique issues and requirements from various business operational processes. It is, however, highly recommended that every effort is made to align to a standard framework solution. This will greatly reduce the overall cost and allow for the business to maximise re-usability across all the business units and departments that use it.
Based on defined and documented methodologies and principles, there must be a strategy in place that will assist the project to stay within scope from a business requirement and technical solution perspective.
Some of the strategies should include: formal project proposal and acceptance by businesses; organisational impact of the project should be analysed and quantified; a data warehouse solution architect must take charge of the overall system designs to ensure all projects are brought in line with the proposed methodology; detailed project plans should be drawn up and remain fixed for the lifetime of the project and must only change under strict change management with proper motivation; business requirements documents that include business objectives, business drivers, key measurements, report specifications and success criteria; and finally, the establishment of an information delivery competency.
One of the most important strategies is that throughout the methodology life cycle - meta-data capturing, data quality assurance and adherence to the methodology must be actively enforced by a participatory architecture and data governance forum.
Planning phase
Project initiation should begin with the establishment of these three technical areas, together with source system measure descriptions, business process descriptions and reporting requirements, in order to capture the detailed scope. This will also ensure very strict change control, signed-off detailed business and technical requirements, mutually understood acceptance criteria and a mutually accepted user testing strategy.
Above all, the size of the project must never be underestimated, and business users should be engaged throughout the delivery design phase, and be involved in review sessions throughout the whole project life cycle. If this is not done, then scope issues can easily derail the whole project or bring it to a screeching halt. Lastly, it is important to use project managers and business analysts that thoroughly understand BI and its engagement.
In essence, developing a project scope and setting proper project expectations creates a starting point for a successful BI implementation strategy, which will ultimately prove to have a positive effect on businesses, and consequently have a fundamental impact on the world of business intelligence.
Share