While working on the Benchmark tool, we wanted a way to visualize the results. Visualizations would allow us to view benchmark results. We could compare results. And view performance over time. We decided to explore Power BI as a means of visualizing the results.
Power BI is a reporting/analytics tool provided by Microsoft. It allows you to pull data from multiple sources, manipulate this data, and create visualizations. I will discuss the pros & cons in the following sections.
Power BI has an impressive list of data sources. There are hundreds to choose from. Supported sources include file based sources, Database, and Cloud. It makes for a really seamless integration with your chosen source. If you plan on working with Power BI, your first step is to ensure your data source is supported.
Once you have selected your data source, the next step is manipulating the data. This can be with an initial query to refine the data. Followed by a UI to model and manipulate the data. This is done in steps, allowing you add, remove, insert steps as needs be. You can name these steps. I would recommend doing this so when you come back to the query later, the steps are self explanatory.
Data can be manipulated in a number of ways. The value type of a column can be changed, for example alphanumeric to numeric. You can write your own functions to create custom columns. Again, this could involve applying formulas, or even just merging data into one column. A manipulation I regularly used was converting a timestamp to date.
Overall, this whole process is hassle free.
Visualizations are pretty simple. You select the data you want to display. Then the visualization you want to display, e.g. pie chart, bar chart, etc. It may take some tweaks in terms of correctly configuring X&Y axis, legends, etc, but ultimately you should have some visuals in mere seconds.
If you have multiple visualizations, they are all responsive to each other. For example, if you refine one report by a particular value, then all other reports will react to highlight the same value. This is a really nice feature.
Publish to Cloud
When you are finished with a report, you can publish it to the cloud. This allows you to access the report from any browser. You can still add/edit visualizations in the cloud. However, you cannot edit the dataset.
Note: now is a good time to mention that you have a limited data allowance in the cloud. Power BI has a data limit of 10GB per user. This is refined further that a dataset can at max be 1GB. This is a big consideration to determine if Power BI is for you.
It’s not free
Power BI is free…to an extent. All of the above in the Pros section is free. So you can do an awful lot before having to upgrade to the paid version. There are two main features that will force you to a paid version. The first one is the ability to share reports. The free version does not allow users to share their reports, or have other reports shared with them. This is a big gotcha, as more often than not you will be creating reports to be shared with others.
The other main perk of the paid version is alerting. You can configure alerts on your reports. So if a certain target is achieved in your report, you may want to be notified about this. This is only available in the paid version.
At time of writing, the cost of Power BI Pro is 8.40 Euro per user per month.
Power BI Desktop – Windows Only
Power BI Desktop is where you will do the grunt of your work. This is where you connect to a datasource, and manipulate the data. Once the datasource is configured, you can work with it online. However, any changed to the model will require it to be edited using Power BI Desktop. At time of writing, Power BI Desktop is a Windows Only application.
Data in Power BI does not automatically sync with updates in its data source. You either have to manually request an update. Or you can configure Schedules to refresh data. You can configure up to 10 scheduled refreshes in a day.
Note: Power BI does have an API to allow you to push data to your reports. Data does update automatically as requests are received. In this scenario, the data is saved directly in Power BI.
Power BI is a powerful tool. It is well integrated with numerous resources. Manipulating data and creating visuals is straight forward. If you do need help, it is well backed online with plenty of tutorials and documentation. For me, the big black mark is on sharing reports. It is unfortunate that both users need the Pro license to be able to view/share reports. Overall, Power BI is worth a punt for your reporting needs. Hopefully this post will help you decide if it is for you.