Power Bi is Microsoft’s maiden attempt in the free, self-service data analysis space. Over time, Power Bi has become a must-have tool for those who are serious about a career in data science and analytics. Let’s learn Power Bi tutorial.
In this article, we provide the ultimate Microsoft Power Bi tutorial; complete with important information including Data files, Power BI Desktop vs. cloud service, Personal vs. Enterprise gateways,12 important Power BI visualization concepts, and Data cleaning via scripts.
What is Power Bi and what are its advantages?
To begin with, Power Bi has all the basic data wrangling features, just like MS Excel’s Power Query. However, it goes several steps further. Its features are multi-fold.
You can build interactive visualizations, type natural language questions to get relevant analysis on your data and create dashboards and reports with a few simple clicks or drag-and-drop. Another advantage is that unlike MS Excel, Power Bi can be used to handle much larger data sets.
Another major advantage of Power Bi is that it can work with almost any data type from Excel to CSV to Access. Not only that it also accesses data from tools like Google Analytics, MailChimp, GitHub, QuickBooks, Salesforce, and others.
The best part is that Microsoft Bi can R scripts. So if you can pull in data via R, you can import it into Power Bi and conduct advanced analysis.
Currently, Power Bi can be accessed both as a downloadable desktop program as well as an online cloud service. While they do have overlapping features, they are not identical in function by any means.
For instance, data wrangling can only be done using the desktop program while dashboards and report sharing can only be done on the cloud platform. However, you can create visualizations and dashboards on either of the two. Power Bi also has mobile apps for iOS, Android and Windows which you can use to view your Power BI reports and dashboards.
Another great thing about Power BI is that it is largely free to use. It does have a paid version which you need to pay $9.99 per month. The only major benefits are increased data storage (10GB vs. 1GB), creation of enterprise “content packs”, higher streaming capacity, and better-timed data refreshes.
Keep in mind that you need a work id to use the software and a regular Gmail id won’t work. You also need to have a Power Bi cloud account to access their free mobile apps. The Power Bi Desktop is, of course, completely free to use and doesn’t require an account, credit card, or even an email address.
Microsoft Power Bi Tutorial
In this section, we provide a bird’s eye Microsoft Power Bi Tutorial containing several key functions. For a more detailed Power Bi tutorial for beginners, check out this video.
It’s one of the best Power Bi tutorial videos out there.
Let’s start with the first important function: Importing Data.
Importing Data into Power BI
Unless you already have data in an actionable format, begin by using the desktop version. That’s because data wrangling is only possible in Power Bi Desktop.
Of course, you can also do your data wrangling using R or Python and then go directly to Power Bi in which case you can use either the desktop version or the cloud-based one.
To import data, go to the home tab and click on the Get Data button. Choose your data source type and then click Connect.
For loading a file, go to Get Data and then select the data type. Choose the file you need to upload and click on Load. Remember you can see a preview of your data before you click load. If there are any issues with the data, click on edit. Use the Microsoft Power Bi editor to make the necessary changes.
It’s always useful to check if number columns are loading in as numeric or text. If you see that numbers are aligned to the left, it means they’re loading as text in which case you need to open the Query editor and change the structure.
In the editor, you can right-click a column header and then click on Change type to change the data type to whole number/decimal number/date, etc. There’s a number of other things you can also do with data wrangling.
Visualizations
Once you’re done with the data wrangling, close the Query editor and then go to the main Power Bi application. Make sure you keep saving your project as you go so that you don’t lose precious work.
Here’s what the Power BI canvas looks like:
Say you’re mapping airline data and want to graph departure delays. If you click on the checkbox next to DEP_DELAY, you’ll get a bar graph which summarizes all the delay time in the data. If you click on Airline then the graph will become a bar graph with total flight delays by the airline.
Keep in mind that unless you specify which format you want the data in, Power Bi tries to guess which graph will be the best visualization tool for that data.
The total delays can, of course, be a factor of the total number of flights an airline has. The more the flights, the higher the total delay time.
In this case, looking at the average or median delay will be more useful. All you need to do is go to the visualizations panel find DEP_DELAY undervalue and find the triangular drop-down menu. Change the Sum of DEP_DELAY to average or median. Here’s what it looks like:
Say you want to sort it from the airline with the highest average delay to the lowest. Go to the ellipsis at the top right and choose Sort By> Average of DEP_DELAY. Here it is:
These are the most basic visualizations available in Power Bi. You can watch a number of Power Bi Tutorial videos to look at all the different visualizations that are possible. Good data science or analytics course will also have a comprehensive Microsoft Power Bi tutorial for beginners.
Power BI Desktop vs. Cloud Service
It can be confusing initially to understand when to use Power Bi desktop vis-a-vis Power BI’s cloud service. Let’s use this power Bi tutorial for beginners to clear up some of the confusion.
For starters, you can only use the desktop version if you have a Windows PC. So if you have a Macbook, the cloud service is your only option.
Thankfully, the cloud service runs on many browsers including Chrome, Safari, Firefox, Internet Explorer, and Microsoft Edge. Mobile apps are also multi-platform and available on iOS, Android, and Windows 10. Of course, apps can only be used for viewing dashboards and reports.
The major advantage of the desktop app is that you can use it to do data wrangling, data modelling, merging different data sources. You can also use it for visualization of course.
The cloud service is more for sharing and creating dashboards. Features like natural language questions using Cortana natural language technology are also available exclusively on the cloud service. You can build visualizations and reports in the cloud service as well.
If you have access to both, then start by using the desktop app for data wrangling and then move to the cloud-based service for creating and sharing dashboards. If you don’t have access to the desktop app (say you have a Macbook or a Linux PC), then you may have to do the data wrangling using R or Python and then go directly to the visualization and dashboards on the cloud service.
Personal vs. Enterprise Gateways
Power Bi allows you to refresh data from local data sources through two gateways; personal and enterprise. Both these gateways require a paid Power Bi Pro account.
If you want to refresh local files on your system automatically, use the personal gateway. If you want to do schedule refreshes, then make sure your computer is on and not in sleep mode.
The enterprise gateway, on the other hand, is designed to be installed on a server but can also be installed locally. With the enterprise gateway, the IT team can manage access for different users.
Conclusion
Power Bi is a relatively recent but extremely powerful analysis and visualization tool. If you already have a career in data science and want to learn Power Bi, there are any number of excellent Power Bi tutorial videos available.
If, on the other hand, you want to get started with a career in data science, you should go for a more comprehensive data science course. In either case, Power Bi is one tool that is a must-have in data analysis; it combines extreme simplicity and intuitiveness with powerful visualization tools.