The Successful BI Project – Start Small … Think BIG
Embarking on a Business Intelligence project can be intimidating.
For many organizations, it’s new territory – a major step beyond simple reporting. A BI project shares many of the same characteristics as more traditional I.T. projects. However, there are aspects of the BI project that can be leveraged to help make the process and deployment a success.
Similar to an I.T. project, the BI initiative requires the support and commitment from the C-Level corner office. The importance of the project to the organization needs to be stressed from above. Like an I.T. project there will be a project team composed of the key stakeholders and subject matter experts (SME’s). It’s important that the individuals on the team not only bring their knowledge and experience to the project, but also have the authority to make decisions on behalf of the users in the areas that they represent. And just like the typical project, there will be regularly scheduled status meetings and updates to the overall project timeline.
A key characteristic that differentiates the BI project from a traditional I.T. project is that with the IT project, we’re typically automating a known process – taking an order, applying cash, performing an inventory adjustment transaction, generating a report. With BI, we’re providing the data repository and the application / infrastructure to the user. At the extreme, what the users do with that data is not part of the project. Yes, we will probably have examples of the types of reports/analysis that the user community is interested in. But it’s the users that will ultimately determine their needs.
An aspect of the BI project that can be leveraged to our advantage during the implementation is the fact that it is an on-going or iterative process. Unlike the more traditional I.T. project, the entire BI infrastructure does not need to be in-place to provide benefit to the organization. Start with one functional area – typically companies start with the Sales organization – and then move on to the next area. By starting with a focused area, you can achieve some of the following benefits:
- Provide results sooner rather than later. Nothing helps to prove the value of the project more than concrete results that are providing value to the organization in a short period of time. Build on incremental successes.
- Enables the end user community and technical staff to obtain a better understanding of the application and the data extract/transform/load process at an early stage. Lessons learned can be applied to subsequent phases.
- Others in the organization will begin to see what BI can do for their functional area. Imagining how BI can provide value is good, but actually seeing it in use is invaluable to deployment of the application in other areas.
- Begins the process of reviewing the condition of the data. Successful data analysis requires data that is of a known quality. Data management and business processes can be put in place to ensure data quality and start to identify other areas in the organization where data quality issues need to be addressed.
When starting the BI project, keep the following in mind to ‘jump start’ the implementation:
- Use a pre-defined data model as a starting point. Although we all think that our business is unique from other companies, we’re actually more similar than different when it comes to the data that we analyze. We all sell goods and/or services to customers. A place to start is much easier to work with than a blank sheet of paper. It’s easier for folks to critique versus being asked what they want/need.
- Avoid the ‘Analysis / Paralysis’ impasse. When researching what data is required for analysis, keep in mind that you’re not going to get it completely right (sorry). Leverage the iterative nature of the process itself. If down the road you realize that something was missed, make the change and move on. Too many projects get stalled at the starting line because project teams are concerned they might miss something. The pre-defined data model helps prevent this. Customize the data model for items specific to your organization.
- The 80/20 rule applies in BI projects. A BI environment that provides value to the organization does not require access to 100% of the data.
- Your business will change and therefore your information and analysis needs will change. The BI project is never really done.
The value that business intelligence can provide a company in today’s challenging environment is unquestioned. With the explosion of data from a variety of sources, having a BI infrastructure in place to support the analysis needs of the organization is more important than ever before. However, implementing a BI solution can be a daunting task, especially if it’s your company’s first experience in this area. Keeping in mind some of the simple steps outlined and an iterative ‘build on success’ approach, the BI project can provide quick results and long term benefits.