Perform Data Analysis in Power BI

Lab story

In this lab, you’ll create the Sales Exploration report.

In this lab you learn how to:

Create animated scatter charts
Use a visual to forecast values

This lab should take approximately 30 minutes.

In this task, you’ll set up the environment for the lab by signing in to Power BI.

Note: If you’ve already signed in to Power BI, skip to the next task.

To open Microsoft Edge, on the taskbar, select the Microsoft Edge program shortcut.
In the Microsoft Edge browser window, navigate to https://app.powerbi.com.

Tip: You can also use the Power BI Service favorite on the Microsoft Edge favorites bar.
Complete the sign-in process with your organizational (or provided) credentials. If prompted by Microsoft Edge to stay signed in, select Yes.
In the Microsoft Edge browser window, in the Power BI service, in the Navigation pane, expand My Workspace. Leave the Microsoft Edge browser window open.

Get started – Create a semantic model

In this task, you’ll set up the environment for the lab by creating a semantic model. If you’ve already published the semantic model, please move to the next task.

In the Microsoft Edge browser window, in the Power BI service, navigate to My Workspace.
Select Upload > Browse.
Navigate to D:\Allfiles\Labs\08-perform-data-analysis-in-power-bi-desktop\Starter folder.
Select the Sales Analysis.pbix file, and then select Open.

If prompted to replace the semantic model, select Replace it.

This method will create a report and a semantic model. We will only use the semantic model to create a new report in this exercise. This same process could be done with an existing semantic model from a different report instead of uploading new. Also, if you aren’t using the report, workspace best practices suggest you delete the unnecessary file.

Create the report

In this task, you’ll create a live connection to the Power BI semantic model created in the last task, and then create a new Sales Exploration report.

Open Power BI Desktop.

Important: If you already have Power BI Desktop open (from a previous lab), close that instance.

Tip: By default, the Getting Started dialog box opens in front of Power BI Desktop. You can choose to sign-in, and then close the pop-up.
In the Home ribbon, select Get Data > Power BI semantic models.
In the Data hub window, select the Sales Analysis semantic model in My Workspace, and then Connect or double-click to load the semantic model.
Navigate to File > Save and save the file name as Sales Exploration in the D:\Allfiles\MySolution folder.

You’ll now create two report pages, and on each page you’ll work with a different visual to analyze and explore data.

Create an animated scatter chart

In this task, you’ll create a scatter chart that can be animated.

Rename Page 1 as Scatter Chart.
Add a Scatter Chart visual to the report page, and then position and resize it so it fills the entire page.

The chart can be animated when a field is added to the Play Axis well/area.
Add the following fields to the visual wells/areas:

The labs use a shorthand notation to reference a field. It will look like this: Reseller | Business Type. In this example, Reseller is the table name and Business Type is the field name.
- X Axis: Sales | Sales
- Y Axis: Sales | Profit Margin
- Legend: Reseller | Business Type
- Size: Sales | Quantity
- Play Axis: Date | Quarter
In the Filters pane, add the Product | Category field to the Filters On This Page well/area.
In the filter card, filter by Bikes.
To animate the chart, at the bottom left corner, select Play.
Watch the entire animation cycle from FY2018 Q1 to FY2020 Q4.

The scatter chart allows understanding the measure values simultaneously: in this case, order quantity, sales revenue, and profit margin.

Each bubble represents a reseller business type. Changes in the bubble size reflect increased or decreased order quantities. While horizontal movements represent increases/decreases in sales revenue, and vertical movements represent increases/decreases in profitability.
When the animation stops, select one of the bubbles to reveal its tracking over time.
Hover the cursor over any bubble to reveal a tooltip describing the measure values for the reseller type at that point in time.
In the Filters pane, filter by Clothing only, and notice that it produces a very different result.
Save the Power BI Desktop file.

Create a forecast

In this task, you’ll create a forecast to determine possible future sales revenue.

Add a new page, and then rename the page to Forecast.
Add a Line Chart visual to the report page, and then position and resize it so it fills the entire page.
Add the following fields to the visual wells/areas:
- X-axis: Date | Date
- Y-axis: Sales | Sales
In the Filters pane, add the Date | Year field to the Filters On This Page well/area.
In the filter card, filter by two years: FY2019 and FY2020.

When forecasting over a time line, you’ll need at least two cycles (years) of data to produce an accurate and stable forecast.
Add also the Product | Category field to the Filters On This Page well/area, and filter by Bikes.
To add a forecast, beneath the Visualizations pane, select the Analytics pane.
Expand the Forecast section.

If the Forecast section isn’t available, it’s probably because the visual hasn’t been correctly configured. Forecasting is only available when two conditions are met: the axis has a single field of type date, and there’s only one value field.
Turn the Forecast option to On.
Configure the following forecast properties, then Apply:
- Units: Months
- Forecast length: 1 month
- Seasonality: 365
- Confidence interval: 80%
In the line visual, notice that the forecast has extended one month beyond the history data.

The gray area represents the confidence. The wider the confidence, the less stable—and therefore the less accurate—the forecast is likely to be.

When you know the length of the cycle, in this case annual, you should enter the seasonality points. Sometimes it could be weekly (7), or monthly (30).
In the Filters pane, filter by Clothing only, and notice that it produces a different result.

Finish up

In this task, you’ll complete the lab in Power BI Desktop.

Select the Scatter Chart page.
Save the Power BI Desktop file.
To publish the file to your My workspace, on the Home ribbon tab, from inside the Share group, select Publish and then select Select to publish.
Close Power BI Desktop.