Self-Service BI - Some Assembly Required
As a parent, I think about Christmas shopping for my children and finding that cool gift that has the words "Some Assembly Required". My first reaction is to play down the cool gift because I don't want have to deal with the hassle of putting it together. The idea of Self-Service BI starts to remind me very much of that cool gift that can be so awesome but only if you have the skills to put it together.
It seems like over the past couple of years there has been a tremendous emphasis put on being able to deliver solutions that provide "Self-Service" analytics. While this sounds great on one hand, on the other hand it reminds me of the wild west days of desktop data marts. The dreaded point where management comes to IT(Information Technology) and asks for the numbers they just saw in a presentation. At this point you know that IT didn't provide the data but where did it come from? The other more important point is "What is the chance that the numbers are accurate?".
I am very curious to see what "Self-Service BI" means to you. Is it one of the following:
- IT provides data that is cleansed, all calculations, and makes this available in a data warehouse where a dashboard/reporting tool is provided for analytical analysis.
- IT provides data that is cleansed, some calculations, and makes this available in a data warehouse. Analysis is done with whatever tool/tools available.
- IT provides some cleansed data, but the majority comes from each department that needs to add in their own data. Calculations are created as needed for analysis. Each department uses whatever tools are available for doing their analysis.
- Other
If you have the time, please share your thoughts. Thanks!
The views expressed are mine alone and do not reflect the views of Farm Credit Services of America or Frontier Farm Credit.
Helping companies deliver predictable, repeatable results | Senior Client Partner
8 年Great post Mike - curious to see what responses you get. Personally, I think data has to be cleansed by the IT team. Too often to individuals entering/generating the data points are either unaware or simply do not care about the importance of clean, accurate data. As far as what calculations IT should perform, I think that has to vary heavily based on level of analytical skills present in each business unit. In a perfect world, those calculations would be uniform across units, but I don't think perfect exists in this case!