You are here:
Add the Data Analysis Subagent to the Analytics and Visualization template in Tableau Next
The Data Analysis subagent in the Analytics and Visualization template enables conversational analytics in Tableau Agent.
Required Editions
| View supported editions. |
| User Permissions Needed | |
|---|---|
| To update Tableau Next settings: | Tableau Next Admin permission set |
| To build and manage Employee agents: | Manage AI Agents, or, Customize Application |
This topic contains instructions for creating a Tableau Agent in Tableau Next that uses the Data Analysis subagent. This topic is related to the section "Create an Agent with the Analytics and Visualization template" in Set Up Tableau Agent in Tableau Next and Create an Agent from an Agentforce Employee Agent Template.
In Summer '26, admins will need to transition from using the Legacy Agentforce Builder to using the Agentforce Studio app.
In Tableau Next, for orgs on Summer '26 (Patch 11.0), use the Legacy Agentforce Builder to create a new agent.
For orgs on Summer '26 (Patch 12.0), use the Agentforce Studio app to create a new agent.
Confirm that Tableau Agent and Data Analysis settings are enabled in Tableau Next, as described in the section "Activate Tableau Agent" in Set Up Tableau Agent in Tableau Next .
-
From Setup, in the Quick Find box, enter
Agents, and select Agentforce Agents. - Click New Agent.
-
In the Agent Creator page, select the Analytics and Visualization V2
agent template, and then click Next.
If you previously created an agent in a previous version of Tableau Next, you may see an Analytics and Visualization Legacy template and an Analytics and Visualization V2 template in Agentforce Builder or Agentforce Studio. Select the Analytics and Visualization V2 template.
- Under Select your agent's subagents, confirm the Data Analysis subagent is added (if not, add it), and then click Next.
- Under Customize your agent, enter a Name, API Name, Description, Role, and Company. You also can use the default settings and then edit them later. Then, click Next.
- Confirm the Default Language and Tone, and then click Create.
- In the Agentforce Builder page, under Subagents, confirm that Data Analysis is listed as a subagent.
- When you're ready, activate your agent.
- To give your users access to your agent, see Manage Employee Agent Access in the Legacy Agentforce Builder.
-
Verify the new agent in Tableau Next.
- Open a metric or dashboard in Tableau Next.
-
Click the Agentforce button
to open the Agentforce panel.
- In the Select Agent list, select the new agent.
- Ask a question about the data and verify a response is generated.
If a semantic model isn’t configured for use with Tableau Agent, it will let you know and direct you to Analytics Agent Readiness enablement. For more information, see Enable Analytics Agent Readiness. - Optional: Set up Agent Scoping to specify a fallback list of semantic models to serve as context for the agent in situations where the context is ambiguous. Click the LLM Model tab to configure the model mode as Best Performance (offers faster responses) or Best Accuracy (offers enhanced reasoning).
Orgs on Summer '26 (Patch 12.0) must use the Agentforce Studio app to create a new agent.
- From the App Launcher, enter
Agent, and then select the Agentforce Studio app. - On the Agents tab, click New Agent.
- Select the Analytics and Visualization V2 agent template. Enter a
name for the agent and then click Let's Go.
If you previously created an agent in a previous version of Tableau Next, you may see an Analytics and Visualization Legacy template and an Analytics and Visualization V2 template in Agentforce Builder or Agentforce Studio. Select the Analytics and Visualization V2 template.
- Continue to follow steps 4 through 11 from the instructions above.
