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.
- Use the C360 Semantic Data Model in Each of Your Data Spaces
On the updated Setup page, view all of your available data spaces, and install a distinct version of the C360 Semantic Model in any of them. With more than one data space, you can segregate your data, metadata, and processes into categories, such as brand, region, or department, and then enable users to see and work with data specific to their category. Previously, you could only add the C360 Semantic Model in the default data space. - Refine Agent Accuracy with More Business Preferences
You can now add up to 50 business preferences to your semantic model. This gives you more flexibility to define your organization's unique terminology and data logic. While this increase from 20 allows for more granular guidance, remember that preferences add tokens to the agent's prompt. Because larger prompts can increase response times, we recommend prioritizing concise, high-value instructions to minimize latency. - Build and Navigate Complex Models With the Improved Semantic Builder Canvas
Build and manage your data models more efficiently with a modern, flexible authoring canvas. Focus on the logic of your data modeling rather than manual screen adjustments. Existing layouts are preserved, so you can start working immediately. - Use New Behaviors in Some Tableau Semantics Functions
Now certain DATE functions, if given invalid input, return a query failure rather than attempt to fix incorrect logic or return Null. This change ensures data integrity and closer alignment with Data 360 and improves performance. Check existing content to see if you must take action. - Use New Behaviors in More Tableau Semantics Functions (Release Update)
- Control Visibility of Expected Range Alert Insight Type in Metric Definitions
When defining metrics, analysts can choose to turn on or turn off the Expected Range Alert insight type. - Jumpstart Semantic Model Creation with Marketplace Templates
Tableau Next Marketplace semantic model templates provide a faster way to create models for insights. These pre-configured templates include mapped data, data model objects, relationships, metrics, and calculated fields. Select from templates focused on your sales, service, or Customer 360 data to quickly generate insights. - Accelerate Your Insights with Metrics Templates in the Marketplace (Beta)
Tableau Next Marketplace metric templates are pre-configured KPIs for sales, service, or Customer 360 data. They install the semantic models and calculated fields necessary to support the metrics, providing you with instant insights. For example, with the Service Insights metrics template, which installs the Service Insights semantic model, you get metrics with KPIs for average wait time, average time to close, cost to serve, and so on. - Unlock Account Growth or Service-Risk Insights with New Tableau Next Metric Templates (Beta)
Get a quicker start with pre-built composite metric templates for the Sales Insights and Service Analytics Agent semantic models. Installing these metrics provides instant insights into your account growth readiness and service-risk adjusted win scores. You can quickly prioritize account action, assessing which accounts have full-funnel readiness or identifying ideal expansion targets based on a combination of strong win rates and low service escalations. - Create Semantic Models Faster with Auto-Generation (Beta)
Use Data Pro to turn your business logic into a ready-to-use semantic model. Describe your analytics goals in natural language, and Data Pro automatically uses data from your workspace to create the structure you need, including data objects, relationships, and calculated fields for your metrics. Skip the manual mapping and get a model you can review and refine in minutes. - Resolve Similar Meanings to Improve Agent Accuracy (Beta)
Use AI to resolve overlapping data definitions and improve agent accuracy. The similarities feature in Semantic Model Optimization (Beta) identifies synonyms, acronyms, and semantic overlaps to prevent ambiguity. For example, similarities can distinguish between Closed Won Deals Amount and Closed Won Deals Amount by Period, so agents don't misinterpret the data.