Cloud BI solutions
To meet these needs, staff are looking for easy-to-use, readily available solutions that can help them meet these needs efficiently, with as little effort as possible. Often, what companies call “official” solutions are not or do not exist, so employees implement them in a “guerrilla” way, using a tool of their choice. We have seen many cases where they have done these analyses themselves, either manually in Excel or by downloading the free version of Microsoft Power BI Desktop, without consulting the company’s IT managers or specialists. At the enterprise level, these tools tend to be treated according to the old 3Ts analogy (forbidden, tolerated, supported) and very often they are simply put in the tolerated category before they become (even by force) an official enterprise BI platform. It is therefore worthwhile to address these needs in a timely manner and to support staff with expertise and tools to avoid heterogeneous and chaotic enterprise tools.
The start of BI implementation
But how can you get a BI project off to a good start? Let’s start small, focusing on solving a well-defined problem. This gives you a good opportunity to get to grips with the technology and produce a usable solution with little investment. Choose data sources that you have access to and are confident of their accuracy. Choose additional data to the main data source (e.g. production data from a database to design data from Excel) and combine them to produce analyses. It is important to serve a real need, not to perform a randomly selected task for the sake of a test.
Let’s think in terms of complete areas of expertise as a way forward. Build up a model for analysing a specialisation, including the relevant indicators. This model will also be suitable for different, more complex analyses. Clarify the concepts and calculated data used in the model and describe them so that it is clear to everyone what is meant. Then we can start to build a real enterprise data model by incorporating data from related disciplines. We may not be able to do everything so easily, so let’s collect the problems we have encountered that we cannot solve in the current environment. These will be the real needs that we will use to develop our system further.
The professional BI solution
What makes a corporate BI truly professional? There are a few aspects to keep in mind to have a sustainable and valuable BI system. First and foremost, you need to have controlled data sources and calculation methods, and the concepts used in the model and reports must be known and documented. After all, this is a prerequisite for users to have confidence in the data that is displayed. It is also important that the publication is done in an appropriate way, so that the verified reports are available to the authorised users with sufficient up-to-date information.
Not all tasks can be done with your own resources. When building an enterprise BI, there are several tasks for which it is advisable to seek the help of an experienced BI or data warehouse developer. These may include processes related to data extraction, integration, but especially building an analytical model or learning best practices for building BI reports and analyses. It is worth using expert help to get started successfully and avoid pitfalls. A slip-up due to impracticality can set back a BI project for a longer period of time or lead to wrong decisions regarding tools and methods.
Complex BI solutions
Of course, life is often more complex than a simple BI solution can meet. We have often found that extracting and integrating data is much more difficult than a standalone BI tool can handle on its own. There are several reasons for this. The problem could be the resource requirements of the integration itself, or the fact that the source system cannot provide historical data because it is constantly overwriting it. Often the model itself is extremely complex, feeding from many sources with different update frequencies. In these cases, it is necessary to build a data warehouse and data marketplaces that can efficiently perform these integration and data marketplace building steps, and then the BI tool can be imposed on top of this as a visualization interface. In addition to these, it is necessary to keep the methodology we started with to test the integration of new sources, to test new concepts. In other words, it should always be possible to connect new data sources to existing ones in an ad-hoc way, as BI needs are constantly changing, just like the company and its environment.
Send us your contact details and we will contact you as soon as possible, or contact us at
+36 1 204-7730!
Leave the BI implementation to us!
We have many years of experience supporting your BI projects.