Tableau Semantics, the the AI-powered semantic layer, maps your raw data to business-friendly terms and logic, the key to building reusable models and metrics that provide contextual, consistent, trusted insights.
- Build and Refine Semantic Models Faster with Semantic Modeler (Beta)
Build semantic models from natural language descriptions of your business cases to accelerate data modeling and reduce technical complexity. Semantic Modeler automatically scans your workspaces to find relevant data objects, suggest relationships, and draft calculated field formulas to deliver a fully mapped, structured semantic model. - Override Semantic Definitions in Extended Semantic Models
When you extend a semantic model, it inherits all definitions from the base model, including data objects, fields, relationships, and calculated fields. Now you can add more functionality to your extended model by overriding the base model semantic definitions. Any property that is overridden is disconnected from the base model. All other properties are connected and therefore continue to receive automatic updates from the base model. - Inherit and Override Business Preferences Across Extended Semantic Models
Extended models inherit Business Preferences from their base model automatically, so you only need to define Business Preferences once. To adapt them for a specific extended model, edit the inherited preferences. The Business Preferences window marks whether the preferences are inherited or overridden, and lets you revert at any time. Reverting removes all changes made in the extended model. Extended models that already had Business Preference changes before this release are treated as overridden and don't inherit from the base model. - Experiment with New Tableau Next MCP Capabilities Before They Reach General Availability (Beta)
The Tableau Next MCP Beta Server gives you early access to new tools for creating and editing dashboards, semantic models, visualizations, and alerts, using Claude, ChatGPT, Slackbot, or any other MCP-compatible client. Test and plan for upcoming capabilities while you continue using the generally available Tableau Next MCP Server for production work. Both servers can run at the same time in your org. - Move Verified Questions To A New Org Using A Data Kit
Verified Questions are now added automatically with their semantic model when packaged into a standard, unmanaged Data Kit. Installing the Data Kit in the target org ensures the Tableau Agent has everything it needs to function, with minimal manual setup. Previously, admins and analysts had to recreate and reverify Verified Questions in each target org from scratch. - Override Calculated Fields and Metrics in Extended Semantic Data Models
When you extend a semantic model, it inherits all definitions from the base model — data objects, fields, relationships, calculated fields, and metrics. Now you can customize that inherited calculation logic and metric definitions to match your business, while everything you don't override keeps inheriting automatically from the base model. This approach preserves a single source of truth; your extended model stays aligned with the base for shared definitions, and diverges only where you intend it to. - Build and Navigate Semantic Models with Smart Canvas
Author and navigate semantic models of any size with the enhanced canvas in the Semantic Model Builder. Smart auto-layout reflows the canvas in real time, relationship lines never cross, and domain-based grouping and focus mode make even large-object Customer 360 models clear and traceable. Switch to the enhanced canvas using the in-product toggle; the previous canvas remains available. - Manage Organizational Hierarchies in Tableau Semantics (Beta)
Answer hierarchy-based analytics questions like "What's my team's revenue?" by adding Role or Manager hierarchies to your semantic models. With one semantic model definition, your org can use organizational hierarchy capabilities across metrics, dashboards, and visualizations including My Team filtering, drill-down navigation, and data roll-up. - Add Predictive Insights to Semantic Models (Beta)
Predictive Insights brings machine learning predictions directly to your semantic models. Reference an AI model, map your fields as inputs, and generate a prediction for each record in place without exporting data, switching tools, or writing code. Optionally, surface the top predictors and recommendations behind each prediction so your users can see what's driving a result and what to do next. - Validate and Calibrate Agent Responses with C360 Semantic Model Questions and Business Preferences
The C360 Semantic Model now comes with a large set of potential questions that your users might ask Tableau Agent with conversational analytics. In Tableau Next Q&A Calibration, you can view the answers to the questions to verify the agent’s accuracy. With your input, the agent can respond consistently and accurately to semantically similar questions. The C360 Semantic Model also includes Business Preferences that define sales, service, and marketing terms to provide additional context for the agent.