ThoughtSpot automates full platform with new Spotter agents

Going fully agentic

Generative AI (GenAI) tools like chatbots that enabled true NLP-based data analysis and let technical experts automate certain tasks, such as code generation and documentation, were the initial vanguard following ChatGPT's launch, which represented significant improvement in GenAI.

These agents are significant not because they introduce features no one else has, but because they stake out a distinctive approach that ThoughtSpot has been edging toward, which is analytics as workflow rather than just a means of creating analytic artifacts.

Since the middle of 2024, however, agents have represented AI's cutting-edge.

Many enterprises have focused their AI initiatives on building agents trained on proprietary data, so the agents understand the unique characteristics of their business. Meanwhile, data management and analytics vendors have similarly built agents that simplify using complex platforms.

For example, ThoughtSpot competitor Tableau in April 2025  launched Next, an agentic AI-based version of its platform featuring agents for data preparation, natural language query (NLQ) and observability. Similarly, Qlik, another of ThoughtSpot's direct competitors, features agents for NLQ and insight generation.

But while competing vendors are providing customers with agents that help perform certain analytics tasks, ThoughtSpot's aim is to provide users with Spotter agents for every part of their analytics workflow, according to Francois Lopitaux, the vendor's senior vice president of product management.

"The idea is to provide an agentic experience for every user of our platform, that at any time during the analytics workflow, there is an assistant that helps you do the dirty work," he said.

Before building SpotterViz, SpotterModel and SpotterCode, ThoughtSpot observed customer behavior to see who uses its platform and how they use it, Lopitaux continued. Then, based on that information, ThoughtSpot developed the new Spotter agents.

"We wanted to provide an agentic experience for every different type of person using our platform," Lopitaux said. "Now, 100% of our platform is agentic."

Like Farmer, Michael Ni, an analyst at Constellation Research, noted that automating all aspects of the analytics workflow is valuable for ThoughtSpot customers.

"What's significant here is that … end-to-end automation turns BI into an agentic system, cutting time-to-insight and eliminating typical team handoff delays that slow most BI and analytics programs," he said.

Meanwhile, the three processes that slow analytics workflows the most are dashboard development and maintenance, data modeling, and embedding analytics tools in applications, Ni continued. As a result, he noted that ThoughtSpot is providing the right Spotter agents to address users' needs. tools in applications, Ni continued. As a result, he noted that ThoughtSpot is providing the right Spotter agents to address users' needs.

"ThoughtSpot prioritized the moments when cycle time explodes," Ni said. "Modeling, dashboarding, and embedding are where enterprises lose days rather than minutes."

Farmer likewise pointed out that the new Spotter agents directly address the three persistent bottlenecks that slow analytics workflows.

He noted that ensuring proper semantics, data governance and accuracy slow modeling. Meanwhile, analysts spend much of their time arranging visuals, and it is difficult for developers to enable analysis and insight generation in external applications.

"These correspond exactly to the roles of SpotterModel, SpotterViz and SpotterCode," Farmer said. "And these bottlenecks tend to limit adoption, not just productivity. So, [the agents are] right on target for the ThoughtSpot audience."

Beyond directly addressing users' needs, ThoughtSpot's new Spotter agents could help the vendor from a competitive standpoint, Farmer continued.

In addition to Tableau and Qlik, Microsoft Power BI is one of ThoughtSpot's main competitors. Power BI features Copilot, which is an AI assistant that helps users throughout the analytics process. It is not, however, a multi-agent system.

"This release confirms ThoughtSpot’s distinctive workflow approach," Farmer said. "They are carving out their own space. I believe they are struggling somewhat to land this story in the market, but this release will help."

Ni similarly noted that the new Spotter agents represent a temporary competitive advantage for ThoughtSpot.

"With this agentic suite, ThoughtSpot is ahead of most BI vendors in automating the full analytics workflow," he said. "Tableau, Power BI and Qlik have strong AI-assisted exploration, but none offer coordinated agents spanning modeling, visualization, embedding and reasoning-based analysis."

However, while ThoughtSpot is perhaps faster than competitors to automate the analytics workflow, some analytics vendors still have broader platforms providing deeper analysis capabilities than ThoughtSpot, Ni added.

"ThoughtSpot’s integrated agent framework sets it apart, even if the company still trails hyperscalers in breadth and some of the analytics vendors in depth," he said.

Looking ahead

With the new Spotter agents scheduled for general availability early next year, ThoughtSpot's next initiative will be to provide users with tools to build their own agents, according to Lopitaux.

In addition to providing agents that simplify the analytics workflow, vendors such as Qlik and Domo also provide their customers with frameworks for building custom agents. ThoughtSpot is planning to do the same.

"The next thing that we are working on is allowing people to create their own agents," Lopitaux said. "This is going to be huge for us in 2026."

Continued focus on agentic AI is wise, according to Ni. However, beyond SpotterModel, SpotterViz and SpotterCode, he suggested that ThoughtSpot develop an agent that surfaces insights and recommends actions.

"ThoughtSpot's next big move should be a Decision Agent," Ni said. "While the analytics market continues to grow, the [emphasis is on] how to increase decision velocity. ... A Decision Agent wouldn't just answer questions. It would surface recommended actions, present options with tradeoffs, and provide the rationale behind each path."

Farmer similarly noted that despite providing agents that automate the analytics workflow, ThoughtSpot could still add more Spotter agents to aid users.

Specifically, the vendor could add an agent that assesses data quality and another that governs metrics, including explaining lineage and detecting semantic changes.

"These feel important in order to govern the workflows the three existing agents will enable," Farmer said.

Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than 25 years of experience. He covers analytics and data management.