Opportunities and Threats
As our experiences show, in analytical applications, the best way to store information is through a multidimensional database, where each substantive area of the application is represented in the form of OLAP cubes. Currently available OLAP solutions offer high-speed data processing and query response while maintaining scalability and operational flexibility.However, a significant threat to the success of implementing an analytical system is the imprecise definition or complete lack of required output information. Often, analysts, unsure of the management's requirements, tend to include all information from a given substantive area offered by the application in final reports. The high volume of reports and their data overload slow down the decision-making process. Thus, it's only a step away for the intended supporting system to become a hindrance to this process.
So how should Business Intelligence solutions be implemented?
It all depends on how prepared the company is to implement advanced BI-class systems. The effectiveness of implementing BI solutions is a derivative of the company's ability to use information as a strategic resource. Conversely, the degree of utilization of information from BI systems is one of the most important metrics of the maturity of a company's business processes. Therefore, the question of the maturity level of individual business processes or the entire company becomes crucial in assessing both the investments in BI projects, the effectiveness of implementing such solutions, and their subsequent use in decision-making processes.
When implementing BI solutions, we always pay attention to the fact that the quality and operation of analytical applications depend directly on the operation of transactional systems that are their data source and on the quality of the data in the databases of these systems. A successful (or unsuccessful) completion of transactional system implementations directly affects the evaluation of BI solutions. The entire process, from recording elementary data (transactions) to publishing synthetic business information, must be critically assessed. Ultimately, from the perspective of the end user, the evaluation of BI solutions is always perceived through the prism of the operation of the analytical application they work with.
"To BI or not to BI"
In order for decision support systems to meet the expectations placed upon them, they must essentially be designed as dedicated solutions from scratch. The aim of implementing BI tools is to achieve an analytical application that accurately mirrors the selected business process while simultaneously maintaining maximum ease of use. However, implementing IT systems based on the BI paradigm is a process in which not only various, increasingly sophisticated data processing technologies must be used but also, perhaps primarily, modern management and motivation techniques that consequently stimulate real demand for information. In many cases, we are faced with the situation of widespread availability of increasingly cheaper and more efficient BI technologies with an absolute lack of concepts for their use in strategic decision-making processes. Furthermore, the development and implementation of BI solutions often require changes within the company itself. Therefore, it is very difficult to distinguish (in the case of success, of course) between the benefits directly resulting from the implementation of analytical applications and data warehouses and the effects of organizational changes accompanying the implementation of BI solutions. On the other hand, it should be taken into account that implementing decision-making systems requires much more significant organizational and cultural changes in the company than is the case with ERP systems. A good reporting system is only a necessary condition, absolutely insufficient, for building an effective decision support system.
