Salesforce has introduced the beta version of Einstein Copilot for Tableau, an AI-powered feature designed to help users at all levels explore data more easily.

Businesses typically rely on expert analysts to create reports and dashboards that provide insights from data. Dashboards in Tableau are interactive and allow users to explore predefined paths. However, there are times when users don’t find the answer they’re looking for within the dashboard and need to dive deeper into the data. Einstein Copilot for Tableau allows users to explore data on their own, using natural language to query and uncover valuable insights from various data sources, including spreadsheets, cloud and on-premises data warehouses, and Salesforce Data Cloud.

Companies across industries are seeking ways to improve efficiency and decision-making, and AI-powered tools are becoming a key part of the solution. While 83% of CEOs want their organizations to be more data-driven, only 30% of employees say their actions are guided by data analysis. More than 90% of business leaders believe generative AI can benefit their organization. Einstein Copilot for Tableau addresses this gap by making data analysis more accessible. It even suggests questions based on the data and metadata analysis, helping reduce the need for constant updates from data analysts and speeding up decision-making.

The feature also includes the Einstein Trust Layer, which provides strong tools to protect data and limit exposure to third-party models. Unlike other AI tools, the Einstein Trust Layer does not store customer prompts or the responses from the language model, ensuring that customer and proprietary data remains private.

Einstein Copilot for Tableau offers several useful features designed specifically for analytical tasks:

• Recommended Questions: When users aren’t sure where to start with their data analysis, Einstein Copilot can suggest relevant questions based on an automated data analysis. For example, a regional sales manager might be guided to ask, “Are there any sales patterns across product categories?” to uncover seasonal trends.
• Conversational Data Exploration: Data analysis is often an ongoing process. Einstein Copilot allows users to continue exploring their data by asking follow-up questions, keeping the context of their original query. A marketer could ask about the performance of campaigns and then dive deeper by asking about specific regions, products, or channels, as if engaging in a conversation with their data.
• Guided Calculation Creation: Creating metrics and KPIs often requires understanding complex calculations. Einstein Copilot helps users by guiding them through the process, whether it’s extracting specific information from a larger data set or combining multiple metrics with the right aggregations to get accurate results.

The goal of Einstein Copilot for Tableau is to make data analysis easier for everyone, regardless of their technical background. It aims to streamline the process of gaining insights and help businesses become more data-driven. As the tool evolves, users will be able to ask natural language questions, visualize insights, and take action based on those insights to create better customer experiences.

Currently, Einstein Copilot for Tableau is in beta for a select group of customers, with plans for general availability this summer. More details will be shared at the Tableau Conference, taking place from April 29 to May 1 in San Diego.