Salesforce Introduces Einstein Copilot for Tableau

Today, Salesforce announced the beta availability of Einstein Copilot for Tableau, a new capability designed to help users in every role and function explore data with AI assistance. 

Businesses tend to distribute insights from data in reports and dashboards created by expert analysts. Dashboards created in Tableau are visual and interactive, allowing users to adjust scope by exploring predefined guided paths. But sometimes a user doesn’t find the answer to their question in a dashboard. They need the ability to perform their own exploration of the data without the prerequisite for deep analytical training. With Einstein Copilot for Tableau, users can dive deep into their data, utilizing Tableau’s powerful analytical engine through natural language to query and derive rich insights from data sources like spreadsheets, cloud and on-premises data warehouses, and Salesforce Data Cloud. 

Organizations in every industry are searching for efficiencies, better decision-making, and are asking their teams to use new tools that leverage AI.

• 83% of CEOs want their organizations to be more data-driven, but only 30% of workers say their actions are driven by data analysis.
• More than 90% of business leaders say generative AI would benefit their organization.

Einstein Copilot for Tableau makes data analysis accessible to every business user, and even suggests questions to users based on analyzing the business data and metadata, helping reduce the number of change requests and updates needed from data analysts, often resulting in faster data-driven decision-making.   

Einstein Copilot for Tableau also leverages the Einstein Trust Layer, giving business and data teams robust tools to help protect data and limit exposure to third-party models. Unlike other AI orchestration engines, the Einstein Trust Layer does not retain customer prompts or the LLM’s responses. This helps customer and proprietary data remain private.

Following Salesforce’s recent introduction of Einstein Copilot, Tableau’s new AI assistant is designed purposely for analytical use cases. Einstein Copilot for Tableau features include:

• Recommended Questions: How do you know where to start when investigating data? Einstein Copilot for Tableau performs an automated analysis of data and recommends relevant questions, guiding the user to interesting potential areas of insight. For example, a regional sales manager could use suggested questions like “Are there any sales patterns across product categories?” to help discover seasonal sales patterns. 
• Conversational Data Exploration: Getting the answer to one’s first question is rarely enough; great analysis requires asking follow-up questions and finding more detail. Einstein Copilot for Tableau enables users to iterate and refine their data exploration. Users can search for an insight without losing the context of their previous question. For example, a marketer could ask what campaigns are performing best and add follow-up questions about specific regions, products, and distribution channels, as if they were truly having a conversation with their data. 
• Guided Calculation Creation: Computing metrics and KPIs is a key task when analyzing data that often demands learning complex calculation syntax. Einstein Copilot for Tableau guides users through this process. For example, a user can get AI-powered guidance to parse information from a longer string field, such as extracting a city name from a full mailing address, or combine multiple measures with the right aggregations to achieve an accurate result.

Salesforce perspective: “Every employee, in every function, must develop fundamental data skills to be successful in the modern enterprise,” said Ryan Aytay, CEO, Tableau. “Einstein Copilot for Tableau streamlines that skill development, helping anyone become experts at understanding data, and enables everyone in the business to surface insights more quickly with trusted AI. Now everyone’s a data expert!” 

Analyst perspective: “Generative AI has the potential to truly revolutionize how insights are accessed and interacted with across the business, but organizations want assurances that the data can be trusted,” said Doug Henschen, vice president and principal analyst at Constellation Research. “With trust guardrails in place, gen AI can evolve from an ambition to a super-charged tool for business.”

As Einstein Copilot capabilities evolve, customers of Salesforce and Tableau will be able to ask natural language questions of data, visualize insights, and turn those insights into action that creates better experiences for customers. More information will be shared at Tableau Conference, April 29-May 1, in San Diego.

Availability: Einstein Copilot in Tableau is currently available in beta for limited customers, and will be generally available this summer.