For decades, Tableau has been helping organizations visualize and understand their data. Thanks to analysts who foster a data-driven culture, capture business requirements, prepare data, and create content for end-users, this mission is succeeding and will continue to thrive. Data analysis is not a skill limited to those with the title of “analyst.” Many of us use data to make decisions, making it one of the most in-demand technological skills today. The rise of generative AI presents a great opportunity to bring transformative benefits to analytics.
In this context, Tableau introduced a series of AI-powered features in 2024, designed to help more users achieve personalized and contextual data insights faster and more easily, directly within their workflows. Among these new tools, the release of Tableau Agent (formerly known as Einstein Copilot for Tableau) marks an exciting milestone. This tool saves time and makes new Tableau analysts even more productive by delivering faster insights through an AI assistant.
Let’s delve into what Tableau Agent is, how it supports users, and how it has evolved over recent months.
What is Tableau Agent?
Tableau Agent is an intelligent AI assistant embedded within the Tableau product suite, supporting all users—from data explorers to seasoned analysts. With Tableau Agent, users can communicate and interact in natural language: it understands the context of your data and assists in creating visualizations, preparing data sources, and efficiently supporting various points in your workflow.
Tableau Agent is a reliable AI analytics assistant built on Salesforce’s Einstein Trust Layer, inheriting its robust security, governance, and trust capabilities. It is now generally available in Tableau Cloud Web Authoring, as well as in Tableau Prep and Catalog, via the Tableau+ edition.
But what can Tableau Agent do, and how can it assist us?
Document and Prepare Your Data
Preparing a data source and writing calculations in a new language can be challenging, especially for new Tableau users. Once data is prepared, it’s equally important to document the data source thoroughly, ensuring that data fields are well-described and clearly understandable for others.
Tableau Agent simplifies both processes, helping users prepare and catalog their data quickly.
Users of Tableau Prep can describe a calculation in natural language to Tableau Agent and receive a calculation automatically generated to add to their Prep flows. There’s no need to write out detailed calculations—just explain the problem and what you want to achieve, and Tableau Agent will create a valid calculation for you.
Once the data is prepared, Tableau Agent can be used in Tableau Catalog to generate comprehensive AI-powered descriptions for data sources, fields, workbooks, and tables. This enriches metadata and makes the data more accessible and understandable for others.
Kickstart Data Exploration
Tableau Agent is an excellent tool for beginners: you don’t need to know all the technical aspects of Tableau to analyze data and gain insights. Tableau Agent suggests relevant business questions to kick off your analysis. As you explore, you can ask deeper questions to uncover additional details.
In a worksheet, a set of recommended questions based on data source metadata makes it easy for users to begin analyzing their data. From there, Tableau Agent helps users iterate and refine data exploration more quickly. You can ask Tableau Agent to show trends over time or relationships between metrics. You can follow up with additional questions to drill into dimensions or focus on specific time periods, with Tableau insights and visualizations adapting automatically.
Important Considerations
Tableau Agent leverages built-in visual best practices guidance, enabling you to create visualizations using simple natural language prompts while reducing complex, time-intensive tasks. Tableau Agent can also perform straightforward tasks like filtering and sorting data, adding labels to bar ends, or changing encodings from color to size.
It translates natural language descriptions into valid Tableau calculation syntax and vice versa. This means you can ask Tableau Agent to create complex calculations, update existing ones, or describe calculations in natural language. Tableau Agent will even name the new calculations for you, making the Tableau workflow smoother and more intuitive.
It’s important to note that Tableau Agent’s responses may vary. The large language models (LLMs) powering Tableau Agent are non-deterministic, meaning responses to the same query can differ. However, like humans, AI is not perfect and may make mistakes. If you encounter inaccurate responses or insights from Tableau Agent, you can use the integrated feedback button to share your experience and help Tableau improve its accuracy and user understanding.
Discover how to achieve fast and intuitive data analysis
Would you like to learn about all the features of Tableau Agent?
Visualitics Team
This article was written and edited by one of our consultants.
Share now on your social channels or via email:
