Does Power BI need to be this complicated?

[Anonymous]

Not sure where to post this but I'm genuinely a bit confused and very curious

To use Power BI, I'm using Desktop, Service and DAX Studio - three different applications to be able to develop and share reports. I also have to understand / learn M, DAX and SQL to be able to get data, model KPIs and diagnose performance issues (ie. review SQL in DAX Studio).

I'm coming from using another stack where database transformations and data modelling were all in SQL, and BI development and distribution on top of the data warehouse used a single BI tool

Are there advantages that I'm missing whereby having multiple tools and languages is a better solution? 

Many things I like about Power BI, just new to it and curious about these things

[parry2k]

@Anonymous to keep it simple, look at it this way:

- Power BI Desktop is your report authoring tools, it connects to the data source, transform/shape data/create a model/ write calculations (like measures), visualize the data, etc.

and from the desktop, you publish it to Power BI Service, and why?

- Power BI Service allows you to share your content with other users in or outside the organization. Currently, you can edit the reports in Power BI service but you cannot model/transform or change the data source. Also, Power BI Service where you will set the schedule of data refreshes to see updated reports.

These are the fundamentals of these tools. DAX Studio is an external tool and not necessarily required, it is mainly used for performance tuning and understand the measures if they are not performing very well. Like SQL profiler. 

I hope it helps.

I would ❤ Kudos if my solution helped. 👉 If you can spend time posting the question, you can also make efforts to give Kudos whoever helped to solve your problem. It is a token of appreciation!

[Anonymous]

Hi @parry2k - thanks for writing, I appreciate you giving an answer

Honestly, I do understand what desktop, service and dax studio do - am using them every waking hour : )

However, I don't understand why the solution to "create a way for businesses to do reporting" is "have two separate products - one on web, one on desktop - and rely on a 3rd party product to help users understand why the performance of your products seems bad". And also, have users learn M and DAX rather than just use SQL.

I'm sure there's an answer that means M+DAX+SQL > SQL, and Desktop+Service+3P > All-In-One

[edhans]

@Anonymous Understand there are a few different things going on.

• M language is just for ETL. It is for massaging data and modeling it, not analyzing it.
• DAX is for analyzing data, not modeling.

While SQL does a bit of both of these, it isn't really the best at either. SQL works best for Add/Change/Delete. That is why MS added SSAS in SQL Server to handle analysis, and Power BI uses the SSAS tabular model.

I'm sure you know all of this as you are deep into it, but that is why SQL isn't used. Power BI isn't about Add/Change/Delete. It is striclty a query tool on the front end, and a modeling tool on the back end to bring in a Star Schema to the DAX model from dozens of different types of sources. 

So it isn't that Power BI has to be this complicated. It is just different. Having used it now for a few years, I have to ask why is SQL so complicated? 😀

[parry2k]

@Anonymous I think this is great model, Power BI desktop has inbuilt performance analyzer and one can use that, but DAX studio adds extra features, it is like similar to SQL profiler, you can argue why there is another add-on SQL profile product, etc etc.

I don't think you need to talk about DAX Studio if you are using it that doesn't mean it is really required. and lot of people don't even use it and there is Tabular Editor which is another external too, heavily used, you use it? I don't think so, it doesn't mean if you are using it or not using it, make you think that you need multiple tools.

End of the day, you need desktop and service, and keep in mind, it is a very young technology/product, and you cannot compare it with matured products who are there for more than 20 years.

I'm sure with time when this product gets mature, we will see more and more features online like Power BI dataflows allow online data prep and transformation.

My 2 cents.