Microsoft has been pushing for a wider use of Power BI Premium, often at the expense of Analysis Services. Despite this, prior to 2021, Azure Analysis Services (AAS) was a clear winner when compared to Power BI Premium.
Since then, however, the gap between the two has become smaller. Many argue the differences now come down to pricing only.
In this article, we look at Power BI and AAS from the cooperation and teamwork point of view. With less technical differences between the two, the ease of work and individual company processes come to the forefront.
Security is a key consideration in most companies. Generally speaking, in Power BI, developers who build a report are the ones who also implement security features. Some organizations believe it’s fine and they don’t have a problem with this.
Others, however, may want to separate this task and let a database team to do security management work. In this case, the database team will implement security from a database perspective in Analysis Services. This will build the extra security wall between the data and report consumers. Depending on security policies, this may be a preferred option for many companies.
Division of roles and duties
This type of separation is closely related to the previous issue. Some companies push self-service BI initiatives to the full. They want users to be self-sufficient and do modeling, calculations, and visualizations all by themselves.
Others, especially larger enterprises, have a dedicated team that creates data models and custom calculations. Then there is a wide group of users across the company who consume these models and create their own visualizations.
Some factors to consider when separating the duties are internal policies, the level of expertise among team members, and whether a company has developers on the outsource.
Working together as a team in one environment is important for many companies. When it comes to shared modeling environment, Power BI is not as good as other software. Obviously, one can do this, however, it isn’t as effective and easy.
On the other hand, outside of Power BI, a team can easily share an Analysis Services model in Visual Studio.
Many developers prefer Visual Studio, which is a tool designed for a shared environment, to build tabular models. When using Visual Studio, developers have more control over the model and its details. Power BI Desktop just does not provide the same level of details for development as Visual Studio.
For reporting purposes, one can use Power BI models in, obviously, Power BI itself and Excel. This option is perfectly fine for many, as Power BI is one of the best visualization tools in and of itself.
However, some want more. Companies that use Analysis Services can do reporting not only in Power BI and Excel, but also in SSRS, and even Tableau and other BI tools. This could be a case for larger companies that use different reporting tools, but still need the single source of truth (SSOT). In this case, an Analysis Services model can be a company’s SSOT, while different department could use their preferred reporting tools.
Disclaimer: the views and opinions of the author expressed in the article do not necessarily reflect those of Centida.