The last build week. You'll assemble a full interactive dashboard in Power BI — line, bar, pie and map visuals working together, controlled by slicers. This is exactly the product your presentation task is marked on.
By the end of this week you can…
Check and fix data types before building anything.
Combine line, column, pie and filled map visuals on one canvas.
Add slicers and use cross-filtering between visuals.
Apply conditional formatting to a map so colour carries meaning.
Lay out a canvas that answers a set of business questions at a glance.
Power BI creates interactive visualisations and dashboards from multiple data sources. Beyond what you built in Google Sheets, it lets people explore data dynamically through slicers, filters, drill-throughs and cross-filtering between visuals — making trends, comparisons and correlations easier to act on.
● Cross-filtering — the feature Sheets doesn't have
In Power BI you don't only filter with slicers. Click a bar in one chart and every other visual filters to it. Click “Cardiology” in the department chart and the map, the trend line and the pie all redraw for Cardiology alone. Click it again to clear.
11.2 Dashboard 1 — fruit shop sales
The situation. A shop wants to track monthly sales across product categories and regions. Use the Example 1 — Fruit Shop sheet.
Task 1 — prepare the data
Open Power BI Desktop and load the Example 1 sheet.
Check the column data types before you build anything:
Date → Date/Time
Sales, Quantity → Whole Number or Decimal
Region, Product Category → Text
● Wrong data type = broken chart
If Sales imports as Text, Power BI can't sum it and your y-axis will be a count instead of a total. If Date imports as Text you lose the date hierarchy entirely. Fix types first — almost every “my chart looks wrong” problem starts here.
Task 2 — line chart, sales trend
Insert a Line chart.
Drag Date to the X-axis — use the day–month hierarchy for grouping.
Drag Sales to the Y-axis, and rename the label to Total Sale.
Set the title to “Daily Sales Trend”.
Click the empty canvas and explore the canvas settings and wallpaper — this is where your dashboard's background is set.
The date hierarchy. Drop a date field on an axis and Power BI offers Year ▸ Quarter ▸ Month ▸ Day. Use the drill arrows on the visual to move between levels — one chart, four levels of detail.
Task 3 — column chart, sales by category
Insert a Clustered column chart.
Product Category → X-axis; Sales → Y-axis.
Title it “Sales by Product Category”.
Task 4 — slicer, filter by region
Insert a Slicer.
Drag Region into the Field well.
Format the slicer: Options ▸ Style ▸ Tile — tiles are easier to tap than a dropdown when you're presenting.
Test it. Pick one region; both charts should redraw. Then click a bar in the column chart and watch the line chart cross-filter.
● Session 1 recap
Types first, then visuals, then the slicer. Line for the trend, clustered column for the comparison, slicer for control — and cross-filtering comes free.
11.3 Dashboard 2 — MediCare Hospital Network
● The brief
You're a healthcare data analyst at MediCare Hospital Network Malaysia. Management wants to monitor patient admissions and treatments over five years (2020–2024) across several branches. They need to know: which diseases are most common, how admissions trend over the years, which departments are busiest, the demographics of patients, and which states carry the highest load.
Use the Example 2 — Hospital sheet: Year, Month, State, Hospital, Department, Disease, Patients, AvgStay, AgeGroup, Gender.
Five questions → five visuals
Start from the questions, not the chart gallery. Every visual on a good dashboard exists because someone asked something.
Task 1 — line chart, patient trend over years
Visual: Line chart. X-axis = Year; Values = Sum of Patients.
Rename the Y-axis label to “Number of Patients”.
Title: “Patients Trend Over Years (2020–2024)”.
Task 2 — pie chart, patients by disease
Visual: Pie chart. Legend = Disease; Values = Sum of Patients.
General ▸ Title → “Number of Patients by Disease”.
Visual ▸ Detail labels → position Outside, show percentage.
Task 3 — filled map, patients by state
Insert a Filled map. Drag State → Location.
Drag Patients into the Tooltips well so the number appears on hover.
Format ▸ Visual ▸ Fill colours → click fx (conditional formatting).
Choose Gradient, based on Sum of Patients.
Pick minimum → midpoint → maximum colours, e.g. light blue → yellow → red.
● Maps need permission and clean names
Map visuals send location names to Bing to be geocoded — if maps are blank, enable them under File ▸ Options ▸ Security ▸ Map and filled map visuals. And spelling matters: “Pulau Pinang” and “Penang” may plot as two different places, so clean state names in Power Query first (Week 5's lesson, new tool).
Task 4 — column chart, patients by department
Visual: Column chart. X-axis = Department; Y-axis = Sum of Patients.
Customise the title, axis labels and column colour.
Task 5 — slicers and layout
Add a slicer with Year in the field.
Add a second slicer with Gender and Age Group — dragging both creates a hierarchy you can expand.
Set the canvas area so every visual fits on one screen without scrolling.
Test every slicer, then cross-filter by clicking into the charts.
One screen, no scrolling. A dashboard that needs scrolling isn't a dashboard — it's a report. Resize visuals until the whole story fits, and remember Week 7's design checklist: headline first, aligned grid, one meaningful accent colour.
11.4 Practice
Both practices use the same download. Build them in Power BI Desktop and apply everything from Week 7's design section.
● Practice 1 · University performance dashboard
Sheet: Practice 1 — University. Management wants to analyse student performance, attendance and enrolment across faculties and campuses.
Line chart → enrolment trends (Year vs Students_Enrolled, by Faculty).
Bar chart → average GPA by Faculty or Program.
Pie chart → gender distribution.
Filled map → campus locations with students enrolled.
Scatter plot → GPA vs Attendance_% — is there a correlation?
Slicers → Year, Campus, Faculty.
● Practice 2 · Fitness centre dashboard
Sheet: Practice 2 — Fitness. A chain of fitness centres wants to understand members' habits across branches. Answer four questions:
Which exercise types are most popular? Column chart — X: Exercise_Type, Y: count of Member_ID.
Calories burned by gender and age group? Clustered column — X: Age_Group, Y: average of Calories_Burned, Legend: Gender.
Attendance trends over five years? Line chart — X: Year, Y: average Sessions_Per_Week, Legend: Branch.
Does workout duration affect calories burned? Scatter — X: Duration_Min, Y: Calories_Burned, Legend: Exercise_Type; optionally size by Sessions_Per_Week.
Slicers → Year, Branch, Age_Group, Exercise_Type.
● This is your presentation product
Practice 2 is close to what the presentation task asks for: several visuals answering stated questions, on one screen, with slicers that reveal something the overall view hides. Build it properly now and you've rehearsed the assessment.
Key terms & definitions
Cross-filtering
Clicking a data point in one visual filters every other visual on the page.
Slicer
A dedicated filter control placed on the canvas; can be styled as a list, dropdown or tiles.
Date hierarchy
The automatic Year ▸ Quarter ▸ Month ▸ Day levels Power BI builds from a date field.
Filled map
A map visual that shades whole regions by value — a choropleth.
Conditional formatting (fx)
Setting a colour from a value rather than fixing it, e.g. a gradient across Sum of Patients.
Tooltip
The detail box shown when hovering over a data point; extra fields can be added to it.
Canvas
The dashboard page itself — its size, background and wallpaper are all formattable.
Data type
How Power BI stores a column (text, whole number, decimal, date/time); wrong types break visuals.
✅ Quick self-check
1. Your Sales column imported as Text. The most likely symptom is…
2. Clicking a bar in one chart and seeing the map redraw is called…
3. To shade states by patient numbers on a filled map, you use…
4. Your dashboard needs scrolling to see everything. You should…
5. The best starting point when designing a dashboard is…