A Beginner’s Guide to Power BI Desktop

6 min readApr 11, 2022

01 — Introduction

Power BI Desktop is an interactive data visualisation software by Microsoft. I like it because it is free to use. An aspiring data analyst can easily pick up the basics by reading a blog post or watch a YouTube tutorial without mulling over licensing or keeping note of time left on free trials. It is also a natural progression after you learnt the basics of Excel formulas, pivot tables and charts, as the interface is familiar.

As part of the Microsoft Power Platform, if you like this application and want to go deeper, there is also the option of taking the next step — by combining it with Power Apps and Power Automate, you can automate workflows, create custom applications and chatbots to tackle business issues.

02 — Downloading

Downloads | Microsoft Power BI

Whether you're on the go or need to create rich, interactive reports, download the tools you need to build reports and…

powerbi.microsoft.com

There are 2 options on the website, the first takes you to the Microsoft Store. If you download it from here, it updates automatically in the background, the update downloads are smaller as they are incremental, and it includes all supported languages. No administrator rights needed for updates.

Second you can download an executable file. It is portable, however if the version is out of date for too long, you won’t be able to open files with it.

03 — Starting your first project

We will be using the QS World University Ranking data set from Kaggle. However in this small project I will only take the top 200 universities in 2022. I have cleaned up the data set and uploaded it here.

You can import the Excel from here.

You can tick the sheet and choose Load.

On the left, there are 3 panels. The top panel, Report, will display your visualisation. The middle panel, Data, will show the raw data in spreadsheet form. The bottom panel, Model, will show the relationships between different models if you have multiple worksheets connected through different types of joins.

For simplicity, this visualisation will only have one sheet and no relational data.

Under the data panel, you can select the different column headers to find out what format the data is in. Here we see the score data type is Decimal number. This means we can do arithmetic calculations with it later such as finding the average score for universities in different regions.

If we change the data type to Text, we will not be able to use it for such a purpose.

04 — Starting the visualisation

For this practice we will make 4 sample charts in the same page. We go back to the Report panel. we can expand the Filters panel and the Fields worksheet.

First we show you how to filter your data set. We drag the rank_display to “Filter on this page”. Which will apply to all the charts in this page of visualisation.

The below settings lets the data show the top 100 universities only. There are many more ways to filter depending on data type. For example if it is Text type, we can filter by rows containing keywords, rows starting with keywords, etc. For numeric data, we can check less than x, x is not blank, data equals to x etc.

05 — Building charts

For the first visualisation, we use the horizontal clustered bar chart to show number of top 100 universities in each country and rank them from most to least.

Select chart type

Drag the borders to resize it

Drag the country column to Axis and Values.

You can do a basic formatting using “Format your visual”. There are 2 tabs — Visual and General.

Under the Visual tab, you can turn on the Data Label by using the slider button and changing its position to outside end.

Under the General tab, you can change the title, bold it, and make it centered.

On the chart, you can sort how the data is shown. Click on the period symbol, sort axis by descending. This will sort the number of top 100 universities by country with the highest on top.

Next we make a pie chart.

We use Region for both Values and Details.

Under Format your Visual, we will turn off the Legend and change the label contents to All detail labels. We can also add a title and format it.

Lastly we can make a simple table.

We will use Region as our argument followed by score, student_faculty_ratio, faculty_count, international_students. By default, numeric values are grouped by Sum. However we will change the first 3 columns to Average.