Tableau Fundamentals
Tableau is the gold standard for visual analytics. Here's how to build dashboards, create calculated fields, design for your audience, and tell stories with data.
The 30-second pitch that won $2 million
A nonprofit fighting homelessness in Los Angeles needed a $2 million grant from a city council skeptical of their approach. The previous year, they'd submitted a 40-page report packed with tables and statistics. The council skimmed it. The grant was denied.
This time, a volunteer data analyst built a single Tableau dashboard. It showed a map of LA with dots for every shelter bed — color-coded by occupancy rate. A time slider showed how demand shifted by season. One bar chart showed cost-per-person for their program versus emergency shelters: $38/night versus $127/night.
The council chair dragged the time slider to December. The map turned red — almost every shelter at capacity. She clicked a neighborhood. The cost comparison updated. She said, "So for the same budget, you could serve three times as many people?" The analyst nodded.
The grant was approved in 30 minutes.
That's not a chart. That's storytelling with data. And Tableau is the tool that makes it possible.
Spun out of Stanford research on interactive data visualization by Chris Stolte, Pat Hanrahan, and Christian Chabot
Tableau goes public, valued at $1.6B on first day of trading
Salesforce acquires Tableau for $15.7B — the largest data analytics acquisition ever
AI-powered analytics with Tableau Pulse and natural language queries
What Tableau does
Tableau is a visual analytics platform. You connect data, drag fields onto a canvas, and Tableau generates the best visualization automatically. It's designed for speed: the question-to-answer time should be seconds, not hours.
✗ Without AI
- ✗Write SQL to extract data
- ✗Export to CSV
- ✗Open in Excel
- ✗Build charts manually
- ✗Copy to PowerPoint
- ✗Email the deck
- ✗Answer follow-up questions by repeating all steps
✓ With AI
- ✓Connect directly to database
- ✓Drag fields to canvas
- ✓Tableau picks the best chart
- ✓Publish interactive dashboard
- ✓Stakeholders explore themselves
- ✓Follow-up questions answered by clicking
Tableau products
| Product | What it does | Cost |
|---|---|---|
| Tableau Public | Build and publish dashboards to the web | Free |
| Tableau Desktop | Full-featured authoring tool | $70/user/month (Creator) |
| Tableau Cloud | Hosted platform for sharing dashboards | Included with Creator |
| Tableau Server | Self-hosted sharing platform | Enterprise pricing |
| Tableau Prep | Data cleaning and preparation | Included with Creator |
The Tableau canvas: rows, columns, marks
Tableau's interface revolves around three concepts:
Dimensions — categorical fields (Region, Product, Customer Name). They define what you're looking at. Blue pills.
Measures — numeric fields (Revenue, Quantity, Profit). They define how much. Green pills.
The canvas — drag a dimension to Rows or Columns, drag a measure to the opposite, and a chart appears.
| You drag... | To... | You get... |
|---|---|---|
| Region (dimension) → Columns | Revenue (measure) → Rows | Bar chart: revenue by region |
| Month (dimension) → Columns | Revenue (measure) → Rows | Line chart: revenue over time |
| Region → Columns, Product → Rows | Revenue → Color | Heatmap: revenue by region and product |
| Latitude → Rows, Longitude → Columns | Revenue → Size | Map: bubble size = revenue |
The mental model: Dimensions slice the data. Measures quantify each slice.
There Are No Dumb Questions
"How does Tableau know to make a bar chart vs. a line chart?"
Tableau uses Show Me — an automatic chart recommender. It looks at the fields you've selected and suggests the best chart type. A dimension + a measure = bar chart. A date dimension + a measure = line chart. You can always override the suggestion, but Show Me is right about 80% of the time.
"What's the difference between Tableau and Power BI?"
Both do the same thing at a high level. Tableau is considered stronger in visualization and exploratory analytics — it was built for visual analysis from day one. Power BI integrates better with the Microsoft ecosystem and is cheaper at scale. If your company uses Microsoft, Power BI has a natural advantage. If visualization quality and community resources matter most, Tableau edges ahead.
Calculated fields: going beyond drag-and-drop
Calculated fields let you create new metrics and dimensions from existing data — without changing the underlying source.
Common calculations
// Profit Margin
[Profit] / [Revenue]
// Sales Category
IF [Revenue] > 10000 THEN "High"
ELSEIF [Revenue] > 1000 THEN "Medium"
ELSE "Low"
END
// Year-over-Year Growth
(SUM([Revenue]) - LOOKUP(SUM([Revenue]), -1)) / LOOKUP(SUM([Revenue]), -1)
// Days Since Last Order
DATEDIFF('day', [Last Order Date], TODAY())
// Concatenate fields
[First Name] + " " + [Last Name]
Table calculations
Table calculations operate on the data already in the view — after aggregation. They're Tableau's equivalent of window functions in SQL.
| Table Calculation | What it does | Use case |
|---|---|---|
| Running Total | Cumulative sum across rows | Year-to-date revenue |
| Percent of Total | Each value as % of sum | Market share by product |
| Difference | Change from previous value | Month-over-month change |
| Rank | Position within a group | Top 10 salespeople |
| Moving Average | Smoothed trend line | De-noising weekly fluctuations |
Create Calculated Fields
25 XPDashboard design in Tableau
A Tableau workbook contains worksheets (individual charts) and dashboards (compositions of multiple worksheets). The best dashboards follow a predictable structure.
The dashboard layout
Title bar — dashboard name, date range, and 2-3 KPI cards (total revenue, growth rate, customer count). Always at the top.
Filter bar — date range picker, region dropdown, category selector. Let users customize their view without touching the data.
Primary chart — the main story. Largest visual on the page, positioned top-left. Usually a trend line or summary bar chart.
Supporting charts — 2-3 smaller visuals that provide context. Product breakdown, regional comparison, funnel metrics.
Detail section — a table or cross-tab at the bottom for users who want to see the actual numbers behind the visuals.
Actions: making dashboards interactive
Actions are what make Tableau dashboards feel alive. When a user clicks a bar, hovers over a point, or selects a region, something happens.
| Action type | What it does | Example |
|---|---|---|
| Filter | Clicking a chart filters other charts | Click "East" on a bar chart → all charts show East data |
| Highlight | Hovering dims everything except the selected item | Hover over "Laptops" → all laptop data highlighted across charts |
| URL | Clicking opens a web page | Click a customer name → opens their CRM profile |
| Go to Sheet | Clicking navigates to another dashboard page | Click "See Details" → opens a detail worksheet |
There Are No Dumb Questions
"How do I make sure my dashboard looks good on different screen sizes?"
Use Tableau's Device Designer to create separate layouts for desktop, tablet, and phone. Set the dashboard size to "Automatic" or choose a fixed size that matches your most common display. Most business dashboards are designed for 1920x1080 desktop screens, then adapted for mobile.
"Should I add a 'last updated' timestamp?"
Always. Put a small text box in the corner showing when the data was last refreshed. Nothing destroys trust faster than a stakeholder making a decision on stale data because they assumed the dashboard was live.
Design a Dashboard
25 XPStorytelling with data: Tableau Stories
Tableau's Story feature lets you create a guided, slide-by-slide narrative through your data. Each "story point" is a snapshot of a dashboard or worksheet with a caption.
Think of it as a PowerPoint presentation, but every slide is interactive.
The three-act data story
| Act | Purpose | Example |
|---|---|---|
| Setup | Show the current state | "Revenue is $4.2M this quarter" |
| Conflict | Reveal the problem or opportunity | "But the West region dropped 23% — driven entirely by one product category" |
| Resolution | Recommend an action | "Replacing the underperforming product line with our top seller from East could recover $380K" |
This is how the nonprofit won their grant. Setup: here's the homelessness problem. Conflict: shelters are at 98% capacity in winter. Resolution: our program serves 3x more people at the same cost.
Publishing and sharing
Tableau Public — free. Your dashboard is visible to everyone on the internet. Great for portfolio pieces and public data.
Tableau Cloud/Server — private. Only people in your organization can see it. Set permissions per user or group.
Embed — place a live Tableau dashboard inside a website, intranet, or application using an embed code.
Subscribe — schedule email delivery of a dashboard snapshot. Users get a PNG in their inbox every Monday morning.
Alerts — set thresholds. "If revenue drops below $100K, email the sales VP." Data-driven notifications without manual monitoring.
Where Tableau dashboards get published (% of users)
Getting started this week
Day 1: Download Tableau Public (free). Connect the built-in Superstore dataset. Drag fields to build a bar chart and a line chart.
Day 2-3: Build a complete dashboard: 3-4 charts, a filter, and a KPI card. Use the dashboard layout principles from this module.
Day 4-5: Create calculated fields: profit margin, a categorization (IF/THEN), and a table calculation (running total or percent of total).
Week 2: Publish your dashboard to Tableau Public. Post the link on LinkedIn. You now have a portfolio piece visible to any hiring manager.
Build Your First Tableau Dashboard
50 XPKey takeaways
- Tableau is built for speed. Connect data, drag fields, and get a visualization in seconds. Question-to-answer time is measured in seconds, not hours.
- Dimensions slice, Measures quantify. Dimensions (blue pills) define what you're looking at. Measures (green pills) define how much. Drag both to the canvas to create a chart.
- Calculated fields extend your analysis. IF/THEN for categorization, arithmetic for metrics, table calculations for running totals and rankings.
- Dashboard design follows a structure: KPI cards at top, filters below, primary chart top-left (largest), supporting charts alongside, detail table at bottom. Five to eight visuals maximum.
- Storytelling wins decisions. Setup, conflict, resolution. The nonprofit didn't win $2M with a chart — they won it with a narrative that used data to make the case impossible to ignore.
- Tableau Public is free and builds your portfolio. Every dashboard you publish is a hiring signal. Start with Superstore, then use real public datasets.
Knowledge Check
1.In Tableau, what is the difference between a dimension and a measure?
2.Why is a filter action the most important type of dashboard action in Tableau?
3.What is the three-act structure for data storytelling, and why does it matter?
4.What is the advantage of Tableau Public for someone learning data visualization?