Objective: in this section we will explore the report. The aim is to learn how we can fulfill an empty report with enriched graphs.
Now that our data is imported and we know the basic features of the dashboard, we can start making our first visualizations. Let's create a new report: go to the left-hand bar menu and click on the dashboard Canada sales data.xlsx, inside the DASHBOARDS tab. And finally click in the box with the name of the database, as you see in the picture. In spite of clicking in the dashboards tab, what we are doing is creating a new report, because the REPORTS section is empty.
A new empty report will appear
NOTE: notice that two new columns have appeared in our table, Month and Year. This is very useful for making our visualizations. All of this information was in the Date field and Power BI creates them automatically for you. Power BI Desktop doesn't do that for you, because you have many different options of managing and transforming your data.
If you click on the Save button and enter a name, a new report will appear on the left-side menu.
These are the main sections when we create a report:
- Visualizations: where you can choose the most convenient visualization, such as bar char, scatter plot, tables, maps... We will cover almost all of them in this lab. You can also import a custom visual, clicking in the three dots on the right of the last Visualization, which are visualizations created by third party.
- Fields and Format tabs: they are right down the Visualizations. On the Fields tab you can put the values that you want in the axis of the graphs. and the Format tab is used to customize the visualization changing the colors, adding titles, changing the font size, etc...
- Fields: here you will find the database, divided in tables. Each table has a list of columns.
As you can see, you have a visualizations menu, and you the Fields menu. Here you'll have all of your tables with it's respective columns. So let's create out first report!
NOTE: there are icons next to some fields. There are three type of icons, the summatory, which means that it is a numeric field; a world icon, which means that there is data about countries on it and we could use it for making interactive maps; and a calculator icon (not in this dataset). This icon appears when we make measurements in Power BI Desktop.
Maybe the canvas doesn't fill all of the space and your charts are too small. You can customize the canvas size as you want, from the View menu.
Let's create the simplest chart, a table. Go to the Visualizations menu on the right-side and select the table. A new box will appear on the blank page.
So now we can fill the table with some data. Remember that we have all of the columns of our database in the Fields section. So you just have to click in the little empty box on the field that you want in the table, for example, Revenue, and then Manufacturer. Now we can find in the table all the revenue of each manufacturer. You may probably can't see the numbers because of the font size. Go to the Format tab (the brush icon) - General - Text Size. Notice that automatically Power BI makes the total revenue for you, in a new row. You can disable this option if you want.
Let's see how easy is to create a graph. Click on the table that we have just created, and then in the visualizations, select the second one, the stacked column chart. Now we have the revenue in Canada, represented by Manufacturer.
What do we have to do if we want to make another visualization with the same data, for example a pie chart? You just click in the chart that you have just created and then select the pie chart in the Visualizations menu. This is an easy way to compare the revenue of all manufacturers.
Easy right? We can customize it a little bit from the Format menu, for example, turning on the Legend.
Let's do a bar chart. Drag and drop Units and Months in the same place in the canvas, and then in Visualizations, select the Clustered Column Chart. Your report should look like this:
Let's see more visualizations.
In order to have better visualizations of our charts, create a new page, in the plus button:
Now we have a new challenge. We'll try to make a 4 dimension chart in a 2 dimension one. Let's see how we can do that. Select the Line and Staked column chart and fill the fields with the followings:
What have we done here? For each year (X axis), we can see all of the revenue spent in that year (Y axis), which is represented by the whole bar. At the same time, we can compare it with the units sold in that year, represented with the line. And finally, we have a segregation of the revenue of each year, separated by the Segment field, in different colors. We can see which color represents a segment in the legend above.
Here we introduce the Slicer option. This is used to filter our data, so that we can focus on individual fields. In this case, we will focus on each Province. How can we do this? First of all, come back to the first page of the report (the one with a Pie Chart), then go to the visualizations pane and choose the slicer icon. Then click on the Province field.
Now if we click in any of the provinces, all the charts will automatically update with the information of that particular province. You can filter the data with more than one option if you go to Format - Selection Controls and turn off the Single Selection option.
Power BI has also great mapping capabilities. It's quite easy to represent our data in an interactive map, in order to look at the geographical relationships in the data. We are going to use embedded Bing maps in the report, and customize them with our particular data. For example, let's see the units sold by province.
Open a new page, click on the Map visualization and then drag the province in the Location field, and the Units in the Size field. This will show us a map with different shapes of circles, regarding the units sold in the provinces. We can also customize it with a legend. Drag and drop the province in the Legend field. We should see a map like this one:
Now we are going to see how can we plot a Treemap. In the same page, drag and drop the Units and the Manufacturer to the canvas. Then select the Treemap. This is a graph very similar to the pie chart. We can see the proportion of units sold for each manufacturer, sorted by maximum size (up - left) to minimum (right - down).
Why have we mixed this charts? because I wanted to show you that there is another way to filter our data. Try to click on one of the Provinces of the map and see how the TreeMap chart changes with the units sold of every manufacturer, in that particular province. If you customize it a little bit, you should have something like this:
