Continue in 2 seconds

Fundamentals of Analysis

  • December 01 2004, 1:00am EST
More in

Building a Solid Foundation for Your BPM System

There are four phases to any IT project: analysis, design, build and deploy. Each of the four phases has its own challenges, both from a technical and manpower standpoint. When the project is building a business performance management (BPM) system, many of the most contentious issues arise in the analysis phase. If your organization is like most, getting analysis done is like trying to get a five-year-old to eat broccoli. Nevertheless, it has to be done - correctly - or you won't have a solid foundation for your BPM system.

Unfortunately, analysis has become synonymous with requirements gathering. It is that, but it's also so much more. The analysis phase of an IT project is where the underpinning for the entire system is laid. The analysis phase, if done right, will yield a solid system that will serve your organization's needs and grow as the business grows. However, if it's done poorly, or not at all, the system won't meet the needs of the user community, and people just won't use it. Then you'll be right back where you started - with database and spreadsheet chaos instead of solid analytical capabilities.

How do you do it properly? The work in each of the four phases can be divided into three tracks: a user track, a data track and a technical track. Dividing the work this way enables the project team to ensure that each component of the overall system - human, data and hardware - is given equal attention. For each phase, then, it becomes a matter of mapping tasks to tracks.

The tasks for the user track are to analyze the presentation and reporting requirements and determine the business rules and metrics for the system. Presentation requirements center around user interface and report design. This task is critical because the interface is the first impression users will have of the system. Reporting is important because the "hard numbers" that the system presents will determine how much trust users have in the system. Defining business rules and metrics means defining the semantic layer, data dimensions and facts, drill-down paths, security mechanisms and data integrity requirements.

For the data track, the tasks involve analysis of source systems, definition of meta data requirements and strategy, and completion of a preliminary data quality assessment. More specifically, the team should concentrate on developing detailed business and technical meta data requirements and schemas that will be the data about the system's data - its data blueprint. This is also the time to develop criteria to assess your organizational data quality and to develop a data quality remediation road map to rectify data quality gaps. Finally, the project team should begin development work on the conceptual data model for the system.

For the technical track, the tasks involve analysis of the current state of the organization's technology and processes and definition of the technical architecture requirements for the new system. As the project team assesses the current state, its focus should be on the impact of the proposed BPM system on the existing enterprise architecture and operational environment. The next step is to evaluate the technology options for integration. This will consist primarily of defining the requirements for the hardware, system and data architectures, as well as selection of the most appropriate hardware and software technologies for each technical architecture component. This step is also where the ETL (extract, transform, load), reporting and data quality tools are chosen.

The analysis phase really is the keystone phase for the entire BPM project. Therefore, beware of potential pitfalls. One is the urge to skimp on the time, or altogether skip the analysis phase by using previously gathered requirements. Don't do it. The world did not stand still from the time you gathered those requirements six months ago until now. You need fresh analysis. Another pitfall is failing to gain consensus on your requirements. All stakeholders need to agree on metrics and reporting requirements, and data quality issues need to be assessed and addressed thoroughly. Finally, make sure that you obtain strong executive sponsorship from the outset of the 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