Transform natural language into visualizations, calculations, and dashboards. Ask questions like "Show me sales by region" and watch the AI build it for you.
What is Tableau Studio?
AI-powered conversational authoring that brings the power of Claude directly into Tableau Desktop
Tableau Studio is the next evolution of Tableau Desktop, integrating conversational AI to transform how analysts build, explore, and share analytics. Instead of manually building every visualization, you can now describe what you want in natural language and let the AI assist you.
The 5 Pillars of Tableau Studio
🗣️ Conversational Authoring
🧠 Semantic Layer
Leverage Tableau's 20 years of VizQL metadata as a competitive moat. The AI understands your data semantics, field types, and relationships automatically.
💡 Adaptive Insights
Integration with Tableau Insights (Pulse/Chiron) delivers deterministic, statistically validated insights directly in your workflow. Ask follow-up questions on any insight card.
🎨 Vibe-Coded Extensions
Build custom dashboard extensions with AI assistance. Describe what you want and the AI generates the code, creating rich interactive visualizations beyond standard charts.
🛡️ Governed Federation
Publishing gates, compliance checks, and audit trails ensure everything stays governed. AI validates workbooks before publishing, catching quality issues automatically.
Why Tableau Studio?
Traditional BI requires manual work for every visualization. Studio accelerates this by 10x while maintaining governance and quality. You stay in control—the AI is your co-pilot, not a black box.
Your First 10 Minutes with Tableau Studio
A progressive walkthrough of how Tableau Studio works — from opening the agent to building your first visualization
Start here — what most analysts do today
Traditional Tableau Desktop requires you to:
- Manually drag every field to rows, columns, marks
- Remember which chart type fits your question
- Hand-code calculated fields with exact syntax
- Hunt through menus for formatting options
- Start over when you want to try a different approach
The problem: Every visualization requires manual construction, even when you know exactly what you want to see.
Traditional workflow:
1. Think: "I want sales by region"
2. Drag [Region] → Columns
3. Drag [Sales] → Rows
4. Change mark type to Bar
5. Sort descending
6. Format colors
7. Add labels
⏱️ 5-10 minutes per viz
2. Drag [Region] → Columns
3. Drag [Sales] → Rows
4. Change mark type to Bar
5. Sort descending
6. Format colors
7. Add labels
⏱️ 5-10 minutes per viz
↓
With Tableau Studio
Say what you want — the agent builds it
Studio workflow:
1. Type: "Show me sales by region"
2. [Agent builds it]
✅ 30 seconds
2. [Agent builds it]
✅ 30 seconds
Tableau Studio gives you:
- Natural language authoring — describe what you want in plain English
- Context-aware agent — understands your data, remembers your preferences
- Iterative refinement — "make it blue", "add a trend line", "sort by profit"
- Skills library — best practices built in (performance, design, governance)
- Full control — you can always drop into manual mode
Step through — your first session with Studio
Follow along with the interactive steps below to see how the agent works
Step 1
Open the Tableau Agent
The big ideas — what makes Studio different
🤖 The Agent is Context-Aware
It knows what workbook you have open, what data sources are available, what sheets already exist. You don't explain from scratch every time—it builds on what's there.
🎯 Skills Encode Best Practices
23+ Solution Engineer best-practice skills built in. The agent applies performance optimizations, design standards, and governance checks automatically—no need to remember them.
🧠 Memory Makes It Personal
Local memories remember your preferences, team conventions, and project context. The more you use it, the better it gets at helping you specifically.
↗️ Pop Out for Multi-Tasking
Detach the agent into its own window. Work across multiple workbooks without losing conversation context—true multi-tasking for complex projects.
💡 Insights Integration
Tableau Pulse insights appear as cards. Click any insight to open the agent with context—"Why did this metric spike?" turns into a conversation, not a dead end.
📁 Project Directory
Configure a local working directory for your team's skills, memories, and project-specific context. Version control it, share it with your team—everyone gets the same intelligent baseline.
You stay in control
Tableau Studio doesn't replace Desktop—it augments it. You can always drop into manual mode, edit what the agent built, or ignore suggestions. Think of it as a co-pilot that handles the tedious parts so you can focus on the insights.
Hands-On Training
Learn by building. Each exercise shows the goal—you figure out how to get there.
Challenge Format:
These exercises show you the desired output. Your job is to build it yourself using Tableau Studio. Expandable hints are available if you get stuck.
1
Apply Style Guide to Workbook
Goal: Validate agent-driven application of a formal style guide (fonts, colors, layout) across a workbook
Data Source: Superstore (PDS 27511317)
Starting Point: Open blank workbook, connect to Superstore, create a simple chart (e.g., Sales by Category)
Your Task: Ask the agent to apply a style guide to your workbook. Observe how it handles typography, color palettes, and layout conventions.
💡 Hints (click to expand)
- Try: "Apply this style guide to my workbook" with a style guide document
- If the agent can't access a URL, you may need to describe the style guide rules instead
- Check consistency across all sheets after applying
- Note: Style guide URL access is a known limitation
Known Issues:
- Agent may not be able to directly access style guide URLs
- Visual regressions possible with icon/tooltip alignment
2
Remap Data Source
Goal: Test whether the agent can remap an existing workbook to a different data source while preserving viz logic
Data Sources:
Starting Point: Open the Superstore sample workbook (or create visualizations with Superstore)
Your Task: Ask the agent to remap your workbook to use the Coffee Chain data source instead. Observe how it handles field mapping and schema reconciliation.
💡 Hints (click to expand)
- Load Coffee Chain as a new data source first
- Try: "Take this workbook and map it to the Coffee Chain data source"
- The agent should reconcile field names and roles between schemas
- Check for orphaned sheets or broken filters after remapping
Known Issues:
- Datasource duplication bug may occur (creates "Coffee Chain (2)" instead of reusing connection)
- May need to manually delete orphaned "Sheet 1"
3
Insights Validation
Goal: Validate that Tableau Insights integration works and you can ask follow-up questions
Data Source: Superstore (PDS 27511317)
Your Task: Connect to Superstore, enable the Insights extension, and interact with insight cards by asking follow-up questions.
💡 Hints (click to expand)
- Go to Extensions menu and enable "Insights"
- Wait for insight cards to load (may take 10-20 seconds)
- Click on any insight card to open the agent with context
- Ask questions like "Why did this metric change?" or "Show me more detail"
Known Issues:
- Insights require a published data source (local files won't work)
- Requires Tableau Cloud connection with Tableau+ subscription
- Insight cards may not load if you're not signed into the magic site
30-second check-in
How was Tableau Studio?
Your feedback shapes the product — rate your experience with the exercises.
How easy was it to use?
Very hardVery easy
How well did it meet your needs?
Not at allExtremely well
Select audience, exercise, and rate at least one scale to send.
✅
Thank you! Your feedback was recorded.
Setup & Getting Started
Instructions for getting Tableau Studio up and running