Webbo3 Power BI · Cohort 1 · Day 1 of 16
Day 1: The Power BI Desktop Interface and Getting Data
Finding your way round the program, and getting Webbo Oil's workbook into it the right way.
By the end of today you can
- Name the three views and the three panes, and say what each is for
- Connect to an Excel workbook with Get Data
- Use the Navigator to pick the sheets you need, and leave out the one you do not
- Choose between Load and Transform Data, and say why it matters
- Build a first visual and save the report as a .pbix file
Today nothing gets cleaned and nothing gets calculated. Today you learn where everything is, and you get Alhaji Musa Bello's data through the door.
That sounds small. It is not: the choice you make at the door (Load or Transform Data) decides whether you spend next week fixing reports or building them.
1. The Power BI Desktop Interface
Definition
Power BI Desktop is one window with three views down the left edge and three panes down the right. The views change what the middle of the screen shows; the panes are where you pick fields, choose visuals and set filters.
Real-Life Example
A car dashboard. You do not need to understand the engine to drive, but you must know where the speedometer, the fuel gauge and the indicators are before you pull out. Ten minutes of finding things now saves hours of hunting later.
Types: The Three Views
| View | Icon on the left | What it shows | You use it for |
|---|---|---|---|
| Report | A bar chart | The canvas: pages of visuals | Building what the reader sees |
| Table | A grid | The loaded data, table by table | Checking what actually loaded |
| Model | Three linked boxes | Every table as a card, with lines between them | Relationships (Week 2) |
There is a fourth place that is not a view: the Power Query Editor. It opens in its own window when you click Transform Data, and it is where cleaning happens (Day 2).
Types: The Three Panes on the Right
| Pane | What is in it |
|---|---|
| Data (older versions call it Fields) | Every loaded table, and every column inside it. You drag from here onto the canvas |
| Visualizations | The visual types (card, bar, line, table, slicer...), and under them the wells: Axis, Values, Legend |
| Filters | Filters on this visual, on this page, or on all pages |
Types: The Ribbon Tabs Across the Top
| Tab | The buttons you will use most |
|---|---|
| Home | Get Data, Transform Data, Refresh, Publish |
| Insert | New page, text box, shapes, buttons |
| Modeling | New measure, New column, Manage relationships |
| View | Themes, gridlines, Selection pane |
Find your way round
Open a blank report. Click each of the three view icons and say out loud what the middle of the screen changed to. If a pane is missing on the right, bring it back from the View tab.
Picture the car dashboard: three dials on the left that change the view, three switches on the right that change what is on it.
2. Get Data
Definition
Get Data is the door every piece of data comes through. You tell it what kind of source you have (an Excel workbook, a CSV file, a database, a web page), point it at the source, and it shows you a preview before anything is brought in.
Real-Life Example
A travel adaptor. Your laptop charger is the same whether you are in Lagos or London; the adaptor is what changes to fit the socket. Each Get Data connector is an adaptor for one kind of source.
Types: The Connectors You Will Meet
| Connector | Use it for | In this course |
|---|---|---|
| Excel workbook | An .xlsx file: pick sheets or tables in the Navigator | Webbo Oil: PowerBI_WebboOil_Stations.xlsx |
| Text/CSV | One .csv or .txt file | PayLink Nigeria's files (the Solo Project) |
| Folder | Every file in a folder, combined into one table | Optional |
| SQL Server, MySQL | A company database | Level 4 |
| Web | A table on a public web page | Optional |
Connecting to the Webbo Oil Workbook
Head Office sends everything in one Excel file, PowerBI_WebboOil_Stations.xlsx. Download it from the Start Here lesson if you have not already, and keep it in a folder you will not move.
Click path
Picture the travel adaptor: plug in, check the light comes on (the preview), and only then charge anything.
3. The Navigator
Definition
The Navigator is the window that appears when a source holds more than one table, such as an Excel workbook with several sheets. It lists them with tick boxes, previews whichever one you click, and brings in only the ones you tick.
Real-Life Example
A shop shelf. You walk along, look at what is on each shelf, and put only what you need into the basket. Taking the whole shelf "just in case" makes the basket heavy and the till slow.
The 7 Sheets in the Workbook
| Sheet | What it holds | Tick it? |
|---|---|---|
| Cover | A title page for people: company details and a list of sheets | No: it is not a table |
| Sales_2023 | Every sale in 2023, typed by seven station managers | Yes |
| Sales_2024 | Every sale in 2024, just as messy | Yes |
| Station_Master | One row per station: city and region | Yes |
| Product_Master | One row per product, and whether it is Fuel or Lubricant | Yes |
| Attendant_Directory | One row per attendant, with their station | Yes |
| Station_Targets | One revenue target per station | Yes |
Why Leave the Cover Sheet Out
Click Cover in the Navigator and look at the preview: 16 rows of titles, labels and notes, with most cells empty. It was written for a person to read, not for Power BI to count. A sheet is only worth loading when every row is one record and every column is one kind of value.
Click each sheet name in the Navigator without ticking it and watch the preview change. Then tick the six data sheets (not Cover) and press Transform Data. Count the queries on the left of the Power Query Editor: there should be six, one per sheet.
Picture the shop shelf: look along it, pick what you need, leave the leaflet by the till.
4. Load or Transform Data?
Definition
Both buttons appear under the Navigator. Load brings the data straight into the report exactly as it is. Transform Data opens the Power Query Editor first, so you can clean it before it arrives.
Real-Life Example
Load is carrying the market bag straight to the table and serving. Transform Data is washing and cutting at the sink first. For bottled water, straight to the table is fine. For vegetables, never.
| Choose | When |
|---|---|
| Load | The data is already tidy: one row per record, clear column names, no typing mess |
| Transform Data | Anything you have not checked, and anything typed in by people. That is both sales sheets, every time |
Why the Sales Sheets Must Never Be Loaded Straight In
Click path
The StationName column alone holds 271 different spellings of seven stations. Load that, build a chart of sales by station, and the chart has a bar for "Wuse Filling Station", another for "WUSE FILLING STATION", another for " Wuse Filling Station" with a space in front, and so on. Alhaji Musa Bello asks why Wuse appears a dozen times, and the report is dead.
For each sheet Power BI adds its own Changed Type step, guessing each column's type from the first 200 rows. On the sales sheets that guess is wrong, because the Date column is typed three different ways. Leave it for now. On Day 2 you delete that step and set the types yourself.
5. After Loading: Table View and a First Visual
Definition
Close & Apply (top left of the Power Query Editor) loads every query into the report. The tables then appear in the Data pane, and Table view shows their rows.
Load the four small sheets now (right-click Sales_2023 and Sales_2024 in the Queries pane and untick Enable load for today, so they wait for Day 2). The quickest check that the right data arrived is the row count at the bottom of Table view:
Your First Visual
A visual is anything on the report canvas that shows data. The simplest is a Table visual. When you drop a text column into it and then a count, Power BI groups the rows by the text and counts each group. That grouping is the idea every visual in this course is built on.
Click path
Save and Refresh
- Save (Ctrl+S) writes a .pbix file: the report, the model and every query in one file. Save as
YourName_Webbo_Day1.pbix. - Refresh (Home tab) re-reads the workbook and re-runs every step. When Head Office sends next month's workbook, you refresh; you do not rebuild.
Build the Region table visual. Then change Count of StationID to Count of StationName. The numbers stay the same. Why? (Because both count the same 7 rows.)
Picture the market bag: the bottled water went straight to the table; the vegetables are still waiting at the sink for Day 2.
Do This Now
- Open Power BI Desktop and find all three views and all three panes.
- Connect to PowerBI_WebboOil_Stations.xlsx with Get Data, Excel workbook.
- In the Navigator, click every sheet, then tick the six data sheets and leave Cover out.
- Choose Transform Data, open Sales_2023 and scroll the StationName column.
- Untick Enable load on the two sales queries, load the four small sheets, build the Region table visual, and save the .pbix.
Checkpoint
Q1. Which view do you use to check that data loaded correctly?
Table view. It shows the loaded rows of each table and the row count at the bottom.
Q2. When does the Navigator appear?
When the source holds more than one table, such as an Excel workbook with several sheets.
Q3. Why is the Cover sheet left unticked?
It is a title page for people, not a table: its rows are labels and notes, not records.
Q4. Why Transform Data rather than Load for the sales sheets?
Because they are messy: StationName alone has 271 spellings of seven stations in 2023. Loading them straight in would put that mess into every chart.
Tick Before You Move On
- ☐ I can name the three views and the three panes
- ☐ I connected to the Webbo Oil workbook with Get Data
- ☐ I ticked the six data sheets and left Cover out
- ☐ I opened Sales_2023 in the Power Query Editor and saw the mess
- ☐ I built the Region table visual and saved my .pbix
Quick Recap
Report, Table and Model are the three views; Data, Visualizations and Filters are the panes. Get Data connects; the Navigator picks sheets when a workbook holds several. Only tables belong in the model: the Cover sheet stays out. Load is for tidy data; Transform Data is for everything people typed. Table view row counts are your first check that the right data arrived.
Final Scenario
Alhaji Musa Bello sends the workbook and asks, "Can you tell me by Friday which of my stations makes the most money?"
You do not open Excel and start adding. You connect to the workbook, look at the previews, see the station names spelled dozens of ways, and choose Transform Data. You have not answered him yet, but you have already avoided the answer that would have been wrong.
Tomorrow you fix what you saw.
Next, Day 2: the Power Query Editor, and cleaning Sales_2023: Trim, Capitalize Each Word, Replace Values, three date formats, duplicates and impossible quantities.