Loading
Salesforce Enforces New Security Requirements in Summer 2026Read More
Tableau Next
Table of Contents
Filter by (0)Add
Select Filters

          No results
          No results
          Here are some search tips

          Check the spelling of your keywords.
          Use more general search terms.
          Select fewer filters to broaden your search.

            Search all of Salesforce Help
            Search all of Salesforce Help
            Create and Refine Semantic Models with Semantic Modeler (Beta)

            You are here:

            1. Salesforce Help
            2. Docs
            3. Tableau Next

            Create and Refine Semantic Models with Semantic Modeler (Beta)

            Build and fine-tune semantic models faster by describing your analytic goals directly to Semantic Modeler in your workspace. It translates your natural language requirements into fully mapped semantic models and helps refine existing ones. The tool automatically scans your active workspaces to find relevant data objects, create relationships, and generates calculated field formulas. This accelerates your time to insights and removes traditional friction, helping both experienced modelers and less expert users build and refine semantic models without manual configuration. Plus, you can review and adjust all selected data objects and formulas before generating the semantic model.

            Required Editions

            View supported editions.
            Note
            Note

            Semantic Modeler is a pilot or beta service that is subject to the Beta Services Terms at Agreements - Salesforce.com or a written Unified Pilot Agreement if executed by Customer, and applicable terms in the Product Terms Directory. Use of this pilot or beta service is at the Customer's sole discretion.

            The Semantic Modeler streamlines your workflow by providing:

            • A Strong Head Start: Automatically creates a new semantic model with the data objects, relationships, and calculated fields from your descriptions. This eliminates the need to build models manually from scratch, empowering less-experienced users while allowing experts to focus on advanced customizations.
            • Faster Time to Insights: Shorten your model creation cycles by leveraging existing data objects to quickly set up active, analytics-ready environments with minimal overhead.
            • Simplified Ongoing Changes: Allows you to easily adapt, update, or refine your existing semantic layer through natural conversation as your business requirements change, while ensuring all modifications remain governed.

            System Limits and Scope

            • Supported Elements: In this release, the natural language capability allows you to describe your analytic goals and business logic to generate a model framework. You can manually modify the selected and unselected data objects presented by the tool. The corresponding relationships and calculated fields are then added automatically by the system to deliver the final model, which is saved directly to your semantic models list.
            • 100-Table Recommendation Scan Limit: The automated recommendation engine scans up to 100 tables across all of the user's workspace to identify matching data objects.
            • Asynchronous Background Processing: Because data object generation processes quietly in a separate tab, your main workspace remains accessible. You can safely continue working on other active tasks while Semantic Modeler builds your semantic model.
            • Out of Scope: Advanced components including metrics, groups, logical views, and Model Context Protocol (MCP) server links are not supported in this version and will be added in future versions.

            Tips for Describing Your Analytical Goals

            Follow these guidelines when using natural language to build or update your models:

            • Be Explicit and Detailed: The more specific your description, the less the agent needs to guess. Include details about your business context, key metrics, and data relationships. For example: Instead of "create a sales model" say "create a sales model tracking quarterly revenue by product family and region".
            • Specify Your Analytic Goals Clearly: State exactly what you want to track or evaluate. For example: Use explicit phrases like "track store performance analysis" or "quarterly revenue data".
            • Identify Your Business Logic: If you need specific metrics computed, state the business intent directly so the system can draft the formula. For example: "Add a calculated field for Average Sales per product family).
            • Explicitly Name Target Objects When Expanding Models: When refining an existing framework, list the new data objects you want to bring in to ensure the discovery engine flags them. For example: "Include the Case and Contact objects".
            • Set Up Semantic Modeler (Beta)
              Enable your users to build and refine semantic models through natural language by deploying the Semantic Modeler. This setup extends your existing Data Pro configuration with specialized capabilities for semantic modeling.
            • Create a Semantic Model with Semantic Modeler (Beta)
              To create a semantic model, simply describe your analytic goals and business logic in natural language to the Semantic Modeler.
            • Refine an Existing Semantic Model with Semantic Modeler (Beta)
              To refine an existing semantic model, simply describe the changes you need in natural language to the Semantic Modeler within the model builder. Whether you’re adding calculated fields, including new data objects, or updating relationships, the Semantic Modeler helps you make these modifications through natural conversation.
             
            Loading
            Salesforce Help | Article