The following are the 3 steps for creating Power BI Reports: get the data from the source, model the data source, and create reports based on that dataset.
Getting Data
Power BI is able to connect to a variety of sources from a flat file, server-based, and cloud-based data sources.
The data can be connected as a live source or as an extract.
Direct query is good for large datasets or real-time data access; however, querying directly to server-based data sources has the following limitations.
- Enterprise Gateway required
- All model tables must come from same database
- No calculated columns or tables in the model
- No data type changes
The following figure explains the connection window for connecting to the data sources. The server-based source has an extra authentication window.
Cleaning up the data
The data should be cleaned before using in the report. The reports need to have the cleansed data and the unwanted rows should be eliminated to improve the performance of the reports.
Data Model
Data modeling is the process of modeling of data tables, E-R data model is required for transactional processes and dimensional modeling is required for reporting, we can create star schema and snowflake schema using the Power BI data model window. In addition to it, we can also join more than two tables. Table relationships are links between tables of data that allow columns in one table to be used meaningfully in another table
To create the data model, there is the managing relationships icon in the Power BI modeling ribbon. When you click that icon, the create relationships dialog box appears as below. We can specify one-to-many or many-to-many relationships between tables, as per our business requirement. Once the relationship is created, the data from both the tables can be accessed based on the common/key column.
Creating Dashboard
Power BI Desktop is predicated on the concept that a picture is worth many thousands of words. Its charting tools let you create clear and convincing visuals that tell your audience far more than a profusion of figures ever could. It will show you how simple it can be not just to make your data explain your analysis, but to make it seem to leap off the screen. You will see how a powerful chart can persuade your peers and bosses that your ideas and insights are the ones to follow.
A little more prosaically, Power BI Desktop lets you make a suitable dataset into
- Pie charts
- Bar charts
- Column charts
- Line charts
- Scatter charts
- Bubble charts
- Funnel charts
- Waterfall charts
![Power BI Reports]()
The visualization pane in the below figure in Power BI desktop is used to select the chart to be used in the dashboard and the columns to be used in the data pane to create this View.
Once you add the view, the charts will be shown on the dashboard as below. One chart can be easily transformed into another chart by just clicking the respective chart required in the chart pane.
Once the charts have been added in the dashboards, it can have interactivity with end users using filters. The following filters can be used in the dashboard to filter the rows and see some specific values.
Once the dashboard has been created, the final view will look like the below screenshot with charts, titles, and filters.
Thank you!