Key Activities of Traditional BI
Continuing with our series, we will further the discussion we started in our initial Operational BI post with a focus on Traditional Business Intelligence. Traditional BI focuses on understanding data obtained from different sources within the enterprise, frequently centering on high-level metrics and KPIs unobtainable through operational BI. The ultimate goal is to have a single source truth reporting system that would cover any useful existing data. It would address questions such as, What has happened? How is this measured historically / against a baseline? What could have caused it? and How are we doing?
Since the term incorporates a blend of technologies, its typical use cases can include analytical reporting (with focus on KPIs, ratios, metrics, variances), exception alerts (automated agents that act when thresholds are met), scorecards (high-level view of the company with indicator), and ad hoc capabilities (such as filtering, drill-down reporting, and slicing and dicing).
Distinctive characteristics of Traditional BI include:
Some key activities when developing a traditional BI system include the following: Defining KPIs; identifying source systems; designing, implementing, and testing Traditional BI system including such elements as dimensional modeling, data integration, and ad-hoc analytical reporting system.
Depending on the scale of the project, staffing needs vary; however, personnel with BI skills are necessary. It is important to have team members with skills in BI technologies and chosen platform tools. This is essential for progress. Since the concepts are similar, people experienced in one ETL tool may be able to work in another within a certain time frame; however, it is risky because there are always best practices, bugs, and workarounds that might not be obvious. For a medium to large-size traditional BI project, the following blend of expertise is needed: BI architect, ETL lead, data modeler, ETL developers, metadata and report developers, business analysts, database analyst, and QA.
In our next post, we will discuss the Data Mining BI.