Tableau Conference 25: Tableau Redefining BI/Analytics in the Agentic AI Era

The event centered on "data and analytics in the agentic era," with Tableau Next - built on the Salesforce platform - as its flagship. Among the fanfare of their DataFam at their yearly Tableau Conference, Tableau also announced a bevy of new features for Tableau Cloud, Tableau Server, Tableau Desktop, and GA dates for components of Tableau Next, which was initially announced in 2024 at Dreamforce and last year's Tableau Conference.

Why this matters? Salesforce and Tableau are building an AI-native decision layer powered by semantics, embedded agents, and real-time workflows. This marks Tableau’s shift from business intelligence to business orchestration.

Highlights of What Tableau Announced?

Key GA announcements and timelines include:

• Tableau Semantics: Now generally available (GA). Integrated with Salesforce Data Cloud, Tableau Semantics provides centralized metrics, labels, and relationships to support natural language questions and semantic queries for analysts and AI alike.
• Agentforce skills (Data Pro and Concierge): GA in June 2025. These assist in data modeling, prep, and conversational analytics using natural language.
• Agentforce skills (Inspector): Beta in Q2 2025 to monitor key business metrics, get alerts, and actionable insights on KPIs and anomalies.
• Tableau Agent is coming to Tableau Public to guide AI-powered dashboard creation.
• Internal marketplace: GA in Q3 2025 for team-based content and agent sharing.

Tableau reassured its customer base by announcing continued investment in Tableau Cloud, Server, Public, and Desktop. Tableau’s CPO, Southard Jones, said he dedicated over half of his development resources to delivering over 130 features (see figure below) to those platforms in the last 12 months. He committed to the audience and later shared his continued roadmap of investments across the Tableau product lines with analysts. This included announcing the Tableau Blueprint to help its customers lead the change towards an agent-powered future.

The audience cheered new features on Tableau’s current platforms, such as Authoring  Extensions API, which opens up a variety of automation tools and visualization extensions; VizQL Data Services, to allow customers, partners, and community members to integrate Tableau platform APIs into open-source AI frameworks; Tableau Pulse Research Assistant for AI-assisted analysis; dark mode (of course); enhanced accessibility features; and Google Sheets integration. Unsurprisingly, the data analysts’ quality of life features received the biggest applause, particularly items like Recycle Bin, which can restore deleted items.

My POV: Delivering the Operating Model For Data & AI

1. Establishing a Data and AI platform edge: Tableau Next is built atop the Salesforce platform to deliver a vertically integrated stack that spans CRM, Agentforce, Data Cloud, Slack, and now, analytics. Tableau Next fills the critical gap between structured enterprise data and dynamic, embedded decision support to form a platform that standalone BI and SaaS application vendors would struggle to match.
2. Delivering value to all customers: Tableau Next means deeper integration, more value from existing data, and a clear path to embedding AI-powered insights directly into everyday workflows. For Salesforce customers, it makes it easier to get more from their CRM investment without switching tools, streamlining how decisions are made across sales, service, and operations rather than using platforms like Power BI or Looker. For non-Salesforce customers, it provides an analytics platform ready for agentic AI.
3. Sending a message on protected roadmaps: Tableau Next sends a clear message to Salesforce customers: Tableau Next is your go-forward platform for agentic analytics. Equally, for the non-Salesforce Tableau user base, there’s reassurance with a clear roadmap: you don’t need to migrate from its leading Tableau Server or Tableau Cloud offerings to provide a foundation for analytic agents or access AI capabilities- summaries, data exploration, and analytics generation agents- but the agentic-building future is clearly being built with Tableau Next on the Salesforce stack.
4. … but with the need for clarity on which road to take: Tableau has built on-ramps for customers to use some of the new AI and agentic capabilities, such as using Tableau Next by leveraging Published Data Sources from Tableau Cloud and Server platforms and Tableau Public getting Tableau Agent (see the figure below on the Tableau expanded family of products). Still, more communication will be needed over the next year on what platform a customer should use. Over half of the conference attendees interviewed were confused about what product lines had what features, what worked with what, and how to frame decisions on platform choice to use for which use case. Coupled with usage-based pricing, those same customers were uncertain about how to move forward. At the same time, all expressed trust that Tableau would take care of them.

What Data and AI Leaders Should Do Next

As one Tableau partner said on the show floor, “[This is the] first year where it became clear you're going to need something different” - Tableau Partner.

For Salesforce customers: Evaluate Tableau Next as an obvious BI/analytics platform for a path to tighter integration, faster time-to-insight, and AI-native workflows. With Data Cloud,  Agentforce, and Tableau Semantics now natively integrated, Salesforce customers should plan to shift focus from dashboards to adopt conversational analytics, new smart alerting on anomalies and derived insights (e.g., using the Inspector pre-built analytics skill), or interactive insights embedded into CRM workflows. For sales, service, and marketing leaders, this means real-time insights without having to leave the context of their applications. The key is to pilot use cases like churn prediction or pipeline acceleration, where semantic metrics and CRM context already exist.

For Tableau customers not on Salesforce: The message from the conference is both reassuring and directional. Customers can continue to use Tableau Cloud and Server standalone. Tableau Semantics is available now, and in June, Tableau+ customers can also try the Tableau Next AI  features—Concierge, Data Pro, Inspector (Beta). By Fall 2025, Tableau Next will support connections to Tableau Cloud and Published Data Sources. However, there’s a strategic fork ahead: full access to agentic features and capabilities like Tableau Semantics, Agent skills like Data Pro and Inspector, depend on  Salesforce infrastructure (e.g., Hyperforce, Data Cloud). As part of a long-term cloud data strategy, customers must evaluate data architecture and pricing plans to decide if integration with Salesforce services or an alternative platform aligns better with their roadmap.

For organizations considering Tableau, Tableau is a leader in BI/analytics, delivering deep data visualizations and storytelling capabilities with flexible deployment options spanning cloud and on-premise, a vibrant community, and a vision for agentic analytics in Tableau Next.  While Tableau is committed to maintaining its standalone BI platforms, CIO’s and CDO’s should recognize that Tableau’s future foundational components are increasingly tied to the Salesforce Agentic Architecture, with greater reliance on components like the Salesforce Data Cloud for data orchestration across data sources, the Einstein Trust Layer for secure and governed AI interactions, and Agentforce to drive action beyond dashboards. CIOs and CDOs should also pay attention to the emerging usage pricing models for Tableau Next to optimize their analytics investments.

Bottom Line: Tableau isn’t just evolving with better visualizations; it’s being redefined into “question & answer” foundation supporting decisioning building atop the Salesforce Agentic Architecture – both as a standalone BI platform and Salesforce’s AI-native analytics. The longer-term unlock won’t be dashboards: it will be agents and semantics driving real-time, context-aware decisions anywhere. More tactically, in the short and mid-term, Tableau provides multiple choices of platforms and onboarding points.

Based on your maturity and incumbent analytics solution, you have many combinations of solutions and potential future starting points. If you have trouble thinking this through, I would love to speak with you and help you work out your path. There are just too many considerations to put into a short blog.

What stood out to you most? Ping me, or drop your thoughts 👇