Skip to content

10 · Project — Build a Sales Performance Dashboard

This capstone combines every Level 1 module into one deliverable: a multi-sheet, interactive Sales Performance Dashboard for Northwind Retail, built against the Orders table (with the State column from Module 9).

Dataset recap

Table: Orders

Order ID Order Date Region State Category Product Sales Quantity Profit
1001 2024-01-05 East New York Furniture Desk 1200 2 180
1002 2024-01-12 West California Electronics Monitor 450 3 90
1003 2024-02-03 East New York Office Supplies Paper 60 10 18
1004 2024-02-20 Central Texas Furniture Chair 800 4 96
1005 2024-03-01 West California Electronics Laptop 2200 2 330
1006 2024-03-15 East Massachusetts Furniture Bookcase 950 1 -40
1007 2024-03-22 Central Texas Office Supplies Binders 120 8 42
1008 2024-04-02 West Washington Furniture Desk 1100 1 150

Step 1 — Build four worksheets

  1. "Sales by Region" (Module 3/4 technique) — Region on Columns, SUM(Sales) on Rows, bar chart, sorted descending. Expected totals: East 2210, West 3750, Central 920.
  2. "Sales Over Time" (Module 4) — continuous Month of Order Date on Columns, SUM(Sales) on Rows, line chart. Expected monthly totals: Jan 1650, Feb 860, Mar 3270, Apr 1100.
  3. "Profit Ratio by Category" (Module 6) — Category on Rows, the Profit Ratio calculated field (SUM([Profit]) / SUM([Sales])) on Columns as a bar, formatted as a percentage. Expected: Furniture = (180+96-40+150)/(1200+800+950+1100) = 386/4050 ≈ 9.53%; Electronics = (90+330)/(450+2200) = 420/2650 ≈ 15.85%; Office Supplies = (18+42)/(60+120) = 60/180 ≈ 33.33%.
  4. "Sales by State" (Module 9) — filled map, State on Detail, SUM(Sales) on Color. Expected darkest state: California (2650).

Step 2 — Assemble the dashboard

  1. Create a new dashboard sized Desktop Browser. Add a Text title object: "Northwind Retail — Sales Performance Dashboard".
  2. Arrange the four worksheets in a 2×2 tiled grid: "Sales by Region" (top left), "Sales Over Time" (top right), "Profit Ratio by Category" (bottom left), "Sales by State" (bottom right).
  3. Add a global Region filter (Module 5, Section 5): right-click the Region filter card on any one sheet → Apply to Worksheets > All Using This Data Source, then show its control on the dashboard.

Step 3 — Wire interactivity

  1. Dashboard > Actions... > Add Action > Filter: Source = "Sales by Region", Target = "Sales Over Time" and "Profit Ratio by Category", Run on Select. Clicking a region bar should now filter both the trend line and the category ratio chart to that region alone.
  2. Dashboard > Actions... > Add Action > Highlight: Source = "Sales by State", Target = "Sales by Region", Run on Hover. Hovering a state on the map should highlight its region's bar without filtering anything out.

Step 4 — Verification pass (do this by hand, not just visually)

Confirm each of the following against the dashboard you built:

  1. With no filters applied, "Sales by Region" shows West as the tallest bar at 3750.
  2. Clicking the East bar filters "Sales Over Time" so only January (from order 1001, since order 1003/1006 are February/March — recompute: East's orders are 1001 (Jan), 1003 (Feb), 1006 (Mar), so the filtered line should show three points — January 1200, February 60, March 950) and "Profit Ratio by Category" to only East's categories (Furniture and Office Supplies, not Electronics, since East has no Electronics orders).
  3. "Sales by State" shows California as the darkest polygon (2650), and hovering it highlights the West bar in "Sales by Region" without removing East or Central from view.
  4. Deselecting (clicking the East bar again) returns all three linked sheets to their unfiltered totals from Steps 1 and Section above.

How It Actually Works

This capstone dashboard is a good place to see the full query pipeline from earlier modules working together on one canvas: each of the four worksheets still issues its own independent GROUP BY query (Module 1's lesson), the global Region filter (Module 7) prepends a shared WHERE clause to all of them from one connection, and the filter/highlight actions (Module 5) inject or remove additional per-click WHERE clauses and opacity states on top of that:

  1. When the global Region filter is set to all regions, "Sales by Region" runs SELECT Region, SUM(Sales) FROM Orders GROUP BY Region, returning East 2210, West 3750, Central 920 (800+120) — matching Step 1's hand-computed totals and confirming West is tallest.
  2. Clicking the East bar adds a second, action-driven WHERE Region = 'East' on top of whatever the global filter already allows, re-running "Sales Over Time"'s query as SELECT Month, SUM(Sales) FROM Orders WHERE Region='East' GROUP BY Month — only orders 1001 (Jan, 1200), 1003 (Feb, 60), 1006 (Mar, 950) qualify, so the three-point line in Step 2 is a direct consequence of that added clause, not a separate hidden query.
  3. The State map's hover-highlight (Step 3) never adds a WHERE clause at all — it's the cheap, no-requery highlight mechanism from Module 5, which is why hovering California can visually tie back to the West bar instantly, with no re-aggregation of Sales figures underneath it.
  4. Deselecting East (Step 4) removes the action's injected clause, and every affected sheet's query reverts to its unfiltered form — this "removing a WHERE clause and re-running" model is also exactly why the filtered totals in Step 2 can be independently verified by hand against Orders: they're not a Tableau-only artifact, they're the same arithmetic a raw SQL query with that WHERE clause would produce.

Cheat sheet — project checklist

Requirement Verified by
4 worksheets built with correct totals Hand-computed totals in Step 1
Dashboard assembled with title + 2×2 layout Visual check
Global Region filter applied to all sheets Filter control changes all 4 sheets
Filter action: Region bar → Time + Ratio charts Click East, verify Step 4.2
Highlight action: State map → Region bar (hover) Hover California, verify Step 4.3

Exercise

Extend the dashboard with a fifth worksheet: a text table listing Order ID, Product, and the Order Size calculated field from Module 6, sorted by Sales descending. Add it to the dashboard and include it as an additional target of the Region filter action from Step 3, then verify by hand that selecting Central shows exactly two rows (orders 1004 and 1007).