To learn more about Power BI, I decided to work through some of the content from the Create and use analytics reports with Power BI learning path from Microsoft’s Learn platform. These are some of my notes and takeaways from those modules.
Power BI consists of three elements:
The general workflow of Power BI would involve creating a report using the Power BI Desktop, publishing it to the Power BI service, and sharing it so users can consume it using a Power BI Mobile App.
The basic building blocks of Power BI are:
Using these fundamental elements we can create a variety of reports ranging from simple to complex.
Visualizations are things like charts, graphs, or maps; ways to visually represent data. They are a useful way to present data in a context that makes it easier to understand and interpret.
Datasets are collections of data. They can be created from a single source or a combination of multiple sources. Power BI features connectors which make it easy to connect to data sources such as Excel, SQL Server, or Salesforce to extract and filter data for your dataset.
Reports are ways to collect and present your visualizations. They can be a single page or multiple pages.
Dashboards are pages from a report that you share with others. They are collections of visualizations with the important constraint that they fit on one page.
Tiles are what hold visualizations on a report or dashboard. Each visualization represents one tile.
In general, with Power BI you can build datasets that contain the information you need, and then use visualizations to create reports and dashboards that tell the story behind those data.
From the Power BI service you can create apps. Apps are just prepackaged reports you can use. To see some, start by going to the Power BI portal. (If you need to sign up for Power BI, you can find information on how to do that here.)
Once you are on your Power BI dashboard, and easy way to see apps in action is to click “Get Data” in the bottom left of the page:
Follow that up by clicking the “Samples” link, which should take you to a page like this:
Choosing the “Sales and Marketing Sample” will take you to a prebuilt report for showing sales and marketing data:
Before we take a look at Power BI Desktop, you will need to download it. You can find information on that here. With Power BI installed, let’s continue.
When you open Power BI Desktop, you should see something like this:
It should look familiar to anyone used to working with Microsoft products:
Take notice of the Publish button in the upper right hand corner. This is how you will publish your report to the Power BI service after you are done creating it.
To load data into Power BI Desktop, choose the Get Data option in the home ribbon. Power BI has a number of built in connectors to attach to the data source of your choice. If you are planning to import data from Excel, make sure that you format your data as a table first. For a few other tips on how to work with Excel as a data source in Power BI, check out the Microsoft documentation.
Inside Power BI Desktop, you can use the Power Query Editor tool to transform the data you are importing. This tool will allow you to duplicate or edit columns to make your data easier to work with. You will need to Apply your changes for them to take effect.
In addition to transforming your data, you can also combine data from different sources. Working within the Power Query Editor you can add new sources of data with the New Source button. You can choose to merge or append that data to your existing query to combine it with the original source.
One benefit of Power BI is that it lets you define relationships between your data so that it isn’t necessary to flatten it. The Power BI Desktop has a Model View canvas where you can see your data sources. You can also visually create relationships between them (along with other tasks such as removing columns or even hiding whole tables). For more in-depth management, the Home ribbon also has a Manage Relationships button that allows you to create and edit relationships.
Sometimes you might want to create a relationship between two datasets when no common column exists between them. In instances like these, it is possible to create new calculated columns that combine fields in the dataset in order to map them. This can be accomplished in the Modeling Ribbon. This ribbon lets you perform other useful operations, such as adding calculated columns, sorting your visualizations, or otherwise optimizing your data model.
Obviously there is much more to learn about Power BI, but hopefully this has been a useful overview and introduction.