Day 3 Homework — Extend, Automate, Publish

Time: about 45 minutes · The final piece of the course

---

The brief

Your capstone dashboard works. Now make it something the business can rely on — which means one more measure, one more way to interrogate it, and a delivery mechanism that doesn’t depend on you being at your desk.

Work in your capstone file from Module 4.

---

Part 1 — Add a new DAX measure (Day 2 M6)

Add one measure your dashboard doesn’t yet have. The first four work straight on the Dashboard Dash model; the last two need the Boss Level model (raw_tables.xlsx), which has the shipments, payments and INVOICES tables. Pick whichever is most useful for your story:

Measure	DAX	Model
Rolling 3-month revenue	CALCULATE([Total Revenue], DATESINPERIOD('Date'[Date], MAX('Date'[Date]), -3, MONTH))	either
Margin vs last year	[Gross Margin %] - CALCULATE([Gross Margin %], SAMEPERIODLASTYEAR('Date'[Date]))	either
Revenue vs target	[Total Revenue] - [Revenue Target]	either
Share of category revenue	DIVIDE([Total Revenue], CALCULATE([Total Revenue], ALL(PRODUCTS[Category])))	either
Days to deliver	AVERAGEX(shipments, DATEDIFF(shipments[ShipDate], shipments[DeliveryDate], DAY))	Boss Level only
Days to pay	AVERAGEX(payments, DATEDIFF(RELATED(INVOICES[InvoiceDate]), payments[PayDate], DAY))	Boss Level only

Requirements:

1. Format it properly (currency, percentage or whole number — not raw decimals)
2. Verify it by hand on a filtered subset before you trust it
3. Put it somewhere it earns its place — a card with a comparison, or a column in the matrix

Part 2 — Add a slicer-driven visual (Day 3 M1)

1. Add one new visual that answers a question the dashboard currently can’t
2. Add a slicer that drives it
3. Use Edit interactions so the slicer affects what it should and nothing else
4. Title the visual as a finding, not a label
5. Set its alt text

If the page is now crowded, remove something. A dashboard that grows without ever shrinking becomes unreadable.

Part 3 — Make it robust (Day 3 M3)

1. Add a “Last refreshed” indicator so readers know how current the data is:
   • In Power Query: = DateTime.LocalNow() as a custom column on a one-row query, or
   • A measure: Last Refresh = "Data as at " & FORMAT(MAX('Date'[Date]), "dd MMM yyyy")
   • Put it in a small card or text box in a corner
2. Add a contact — your name or team — so people know who to ask
3. Apply row-level security and test it with View as role. On the Dashboard Dash model, filter on branches[City] — e.g. [City] = "Mutare" for a branch-manager role. (The Boss Level model has a ready-made security sheet mapping users to regions if you’d rather map roles to users.)

Part 4 — Publish and deliver (Day 3 M3)

1. Publish to a proper workspace — not My Workspace
2. Configure scheduled refresh, and turn on failure notifications
3. Note which sources would need an on-premises gateway, and why
4. Export to PDF and check it still reads well as a static document
5. Try Analyze in Excel against your published model
6. Create an app, or add the report as a Teams tab

---

If you don’t have a Power BI account

Parts 1 and 2 work entirely in Desktop. For Parts 3 and 4, write a short deployment plan instead — it’s the more valuable skill anyway:

• Which workspace, and who gets which role?
• Who are the consumers, and what licence does each need?
• Which sources need a gateway?
• What’s the refresh schedule, and who gets the failure alerts?
• What are the RLS rules, and how do you verify the identity match works?
• App, direct share, or Teams — and why?

---

Self-assessment

Tick these off honestly:

• The new measure is correct — I verified it by hand, not just by eye
• It’s formatted, not showing raw decimals
• The new visual answers a real question
• Interactions are deliberate, not default
• Titles state findings
• Readers can see how fresh the data is and who to ask
• RLS tested as a Viewer, not as an Admin
• Refresh failure notifications are on
• I removed at least one thing to make room
• I can present the whole thing in two minutes

That last one is the real test. If you can’t, the dashboard is still telling too many stories at once.

---

And that’s the course

You started with a grid of empty cells. You can now:

• Clean and structure data that arrives in whatever state it happens to arrive in
• Model it correctly, so the totals are actually true
• Calculate on it with DAX, including over time
• Present it so a busy person understands it in five seconds
• Publish it so it refreshes without you
• And say what should be done about it

The last one is what makes the rest worth having.

---

Back to: Day 3 index · Course home: README