Secrets of Mastering Data Analytics with Power BI (the concluding part)

Practice Makes Perfect!!!

It's the perfect way to master anything. Something happened one day, during my Banking years. Myself and two of my colleagues for some reason I can't remember, started an argument on who was most involved in playing rough as kids. We were challenging one another with different things we did as kids. We quickly eliminated one person from the contest when we asked him if he used to climb orange trees and he said yes. No way!!! In case you didn't know, orange trees are full of thorns.

When it was time to find out the winner, my finalist colleague rolled up his trousers and sleeves to display numerous scars on his body. I don't have such scars and I told him it's not by scars. I tried to back up my argument, telling him how I used to walk long distance with my hands while my legs would be up in the air. He was very quick to say he also used to do that. Then I asked, "Can you still do that now?" He said no, his "bones are now old". And that was it, because i could/can still do it. He definately didn't believe me until I walked with my hands in the banking hall. Then we both agreed I was not on his level.

That was practice. That was something I used to do over and over again as a kid and i think i'll still be able to walk with my hands when I am 80 years old.

Back to Data Analytics with Power BI

If you have read my first article on this topic (Secrets of Mastering Data Analytics with Power BI (part 1)), then you already know the attitude, behavior and path to take on this journey. This time, my focus is on the technical side.

You must understand that Power BI stands on certain pillars which I will be describing below. You should learn Power BI with these pillars in mind. As a matter of fact, spend time and focus on each of these pillars separately.

Core Pillars of Power BI

1. Power Query & M: Because everything starts with data, you should know how to get and prepare the data you need. A lot of times we get data we term "dirty data" and we end up spending a lot of time cleaning it. Power Query is an excellent tool for that, and the process can be automated. It uses a functional language called "M" which you are not required to know at the beginning as nearly all tasks required of it can be completed by clicking buttons on the user interface. While every button click writes out the "M "codes for you, it makes sense that you start looking at those codes and become familiar with the "M" language for more advanced usage. Start up by reading dbrownconsulting's 7 Golden Rules of Data. Its gives a solid foundation in knowing what a clean data is and how to structure your data before setting out for analysis. You should also check Here to see the Power Query books you can purchase to further your quest.

1. Data Modeling: After getting data, which Power Query would have done for you, data modeling is the next thing. A data model allows you create connections between so many data tables without having to flatten your data like one would do in Excel by writing VLOOKUP() formulas to bring in the missing parts of a data set. Just connect the two tables. Learn about the concepts of Fact Tables and Dimension Tables. A data model is like the foundation of a house, make it solid. A well prepared data sets with a solid data model is what makes writing formulas easy.
2. DAX (Data Analysis Expressions): Take it as a rule not to write any DAX formula until you have clean data sets and a proper data model. DAX is the formula language of Power BI and it is very similar to Excel formulas. You will come across many Excel functions that are also available in DAX. But DAX behaves differently from Excel formulas. You need to understand how the DAX engine works. Once you understand this behavior, writing formulas in Power BI won't be difficult. You should check here for a list of books covering this topic. They also cover Data Modeling.
3. Data Visualization: This is usually the endgame of report authoring. Knowing how to present your analysis is as important as knowing how to prepare data with power query, building a data model and writing DAX. Because data have stories to tell, you should know when to use a table or chart/graph, the kind of chart you need to use and the ones to avoid, use of colors and pre-attentive attributes to tell your story well. I recommend Steven Few's www.perceptualedge.com and Cole Naussbaumer's www.storytellingwithdata.com. You can also get a list of books covering the topic here

Conclusions

This is the data age, everyone should know how to work with it. There are thousands of courses, specialties and tools on data alone. Power BI is leading the pack of BI tools. Following the tips from this article and the preceding one will make a huge difference in the way you work with data. Practice all of them until they become second nature to you.

In my next articles, i will be treating the above mentioned pillars one at a time, explaining concepts and "How To's".

ps: get this free pdf book ( Introducing Microsoft Power BI by Marco Russo & Alberto Ferrari from Microsoft Press Store.