Agentic Experiences25% of the exam
Tableau Agent conversational analytics and calibration
How the Analytics Agent (Tableau Agent with the Data Analysis subagent) answers questions through semantic queries, which models it searches, what context and question types it supports, and how Q&A Calibration (Beta) and Business Preferences tune its answers.
Understand how the Analytics Agent enables users to explore data and generate insights.
~8 min read · 19 practice questions
What the exam expects
- Explain how the Analytics Agent answers a question with a semantic query against the model behind a metric or dashboard.
- Configure Agent Scoping and predict which semantic models the agent searches with and without page context.
- Match a request to supported question types, surfaces and context, and spot unsupported types and PDS function limits.
- Use exact-match quoting, Sources, metadata questions, follow-up chart changes and the Slack thread workflow.
- Run Q&A Calibration (Beta) and write Business Preferences that respect their precedence, scope and format rules.
Key terms:
- Tableau Agent: Help's current name for Agentforce for Analytics; with the Data Analysis subagent it provides conversational analytics (formerly Concierge: Analytics Q&A), which the exam guide calls the Analytics Agent.
- Fallback list: the admin-approved semantic models an agent searches when a question has no clear context.
- Verified question: a calibrated question and answer saved with its semantic model.
- Business Preference: free-text guidance in a semantic model that tells Analytics Q&A how the business uses its data.
Key facts
Query path
- Detail
- Semantic query against the associated metric or dashboard's model
Fallback list editors
- Detail
- Tableau Unmetered Admin or Tableau Next Admin; analysts view only
Context available
- Detail
- Agent uses dashboard or metric models instead of the fallback list
Response provenance
- Detail
- Sources shows analysis details, data sources and metrics
Slack thread answers
- Detail
- Preview first; Share in Thread reveals it
Quoted value
- Detail
- Exact, case-insensitive match
Unsupported question types
- Detail
- Hypothesis testing, root cause, window functions except Period over Period
Verified question limit
- Detail
- 100 per semantic model
Business Preference size
- Detail
- Up to 50 instructions, 300 characters each
Supported hierarchy types
- Detail
- Role and manager hierarchies
Sources for this section (9)
- About Conversational Analyticshelp.salesforce.com (opens in a new tab)
- Agent Scopinghelp.salesforce.com (opens in a new tab)
- Ask Questionshelp.salesforce.com (opens in a new tab)
- Slack Q&Ahelp.salesforce.com (opens in a new tab)
- Supported Questions and Surfaceshelp.salesforce.com (opens in a new tab)
- Limitationshelp.salesforce.com (opens in a new tab)
- Q&A Calibration (Beta)help.salesforce.com (opens in a new tab)
- Business Preferenceshelp.salesforce.com (opens in a new tab)
- Organizational Hierarchies (Beta)help.salesforce.com (opens in a new tab)
- What does the agent do when a quoted exact-match search returns nothing?
- Why can adding many Business Preferences slow the agent?
- When Agentforce isn't enabled, what does the metric details page offer instead?
- Which hierarchy filter does Tableau Next currently offer for organizational hierarchies?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Tableau Next Consultant pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 21 lessons of the Tableau Next Consultant guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (19 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Choose guide-only, practice-only, or save $4 with both