The field of Business Intelligence has developed dramatically in the last decade. More and more organizations are embarking on BI projects of various scopes. In some projects, mistakes are repeated that will ultimately result in an unsuccessful project:
Lack of senior support and business involvement
Everyone knows that a large IT project requires management support, a "tailwind." In BI projects especially, the lack of active involvement from marketing managers, operations managers, and other senior business decision-makers can lead to missing out on system specifications and its implementation and integration. These senior executives are needed for their frequent input of business processes and reflection of the company's strategy to guide the IT implementation team in defining and delineating the central core areas required. BI projects are known to exceed their original allocated budget. This overrun occurs when implementers do not receive defined boundaries from management regarding the scope of critical data required. As a result, databases become monstrous, and parts of them are even unusable.
Lack of proper data examination
A solid technological platform with high capabilities for segmentation and analysis doesn't know whether the data reflected from it is correct or not. Poor-quality data can and often will impair the reliability and benefits of data warehouses and BI systems in general. The challenge in data matching is not only computational but mainly a business challenge. Cleaning and improving existing organizational data and information will ultimately affect day-to-day work and, of course, the BI project.
Unfriendly usage
BI implementers sometimes forget that their development's primary beneficiaries are organization employees from a wide range of fields and different departments. The profile of the average system user is not fixed and can range between very different levels of computer literacy and business understanding. The user interface and query execution should be intuitive. A user who gets "stuck" in their first system use will not always return to it.
Low performance
User expectations for query response time are much higher than thought. If the data volume in the data warehouses is too large for the designated number of users, consider adding more processors or developing more cubes to improve system performance. It's advisable to perform these actions before going live and not after. Another mistake made by the same family is ignoring the initial enthusiasm factor. A successful system generates excellent interest.
Too many or too few tools
IT departments need to carefully examine the number of tools available and accessible to employees. Too many tools and systems lead to confusion and high training and implementation costs, while too few tools lead to frustration and disappointment.
Doing it alone
Ten years ago, finding BI experts for specific markets was difficult. Today, due to the maturity of the field and the large amount of research done on the subject, it's a mistake to embark on a large BI project without thoroughly examining the organization's processes, projects, products, and people. In large organizations, it's even recommended to establish a BI Competency Center, a core body containing internal and external experts focused solely on the project. This focused body will advise business units in the organization, accompany them, and prevent additional mistakes.
Too rapid data growth
One of the central challenges for those engaged in BI work is controlling the data growth rate. Too rapid data growth can lead to chaos in systems, exceed defined core areas, and lead to erroneous decision-making.
Lack of flexibility for changes
Thanks to globalization, the volatile economy, and other factors, building rigid data warehouses that cannot be changed is the sure way to project failure. An organization's strategy and goals can change, additional factors can affect a particular process, and parameters can be added or removed from data warehouses. The system must be allowed design flexibility to adapt it to relevant needs.
Ignoring external factors
BI systems must include data and information from external sources. External factors such as the economy, politics, regulations, and even weather can influence decision-making and the execution of strategic moves in organizations.
Incorrect/non-reflective customer data
If customer satisfaction is the key to your organization's success and you were asked to develop a tool to quantify satisfaction, try to overcome the urge to conduct an annual satisfaction survey. An annual survey won't help when up-to-date information is needed every week or month.
Examine all your existing sources, such as the call center, help desk, etc. - process them and incorporate them into queries.
Comments