Loved by Business Leaders, Trusted by Analysts
Last year, we introduced Spotter our AI analyst that delivers agentic data experiences with enterprise-grade trust and scale. Today, we’re delivering several key innovations that will help you streamline insights-to-actions with agentic analytics, crossing a major milestone on our path to enabling an autonomous business.
Today, we’re introducing new Spotter capabilities that revolutionize the way business users can interact with their data for actionable insights. Let's explore how Spotter’s full range of skills—from answering business questions down to deep analysis and brainstorming—can empower teams to make data-driven decisions using agentic analytics.
Level 1: Basic questions
Any analysis starts off with a question—and before you even have your first question, Spotter gives you some places to start. Whether on your desktop or on your phone, Spotter provides a set of curated questions that help you get an understanding of your data.
Once you click on one of these starter questions (or ask your own), Spotter instantly highlights what's trending up or down, making it easy to drill down to investigate. The built-in AI automatically explains why something is changing without waiting for analyst support or complex SQL queries.
This first level of inquiry provides actionable insights between meetings, during commutes, and anytime when you’re not sitting at your desk, which puts business intelligence right at your fingertips.
In this case, let’s ask Spotter on the mobile app, “What is my churn by age group?”
Level 2: Understanding your dataset
To find connected insights in your business data, you need to first understand what data is contained in the dataset. This is often a challenge for business users who aren’t familiar with the source data. With Spotter, getting a quick overview of your data is as easy as asking, “What questions can you answer?”
Spotter quickly translates your datasets into business-friendly terminology so business users can confidently explore their data through natural language conversations. With its AI-powered interface, Spotter guides users with proactive, relevant questions and column-level context.
Here’s what happens when a customer success representative asks Spotter, “What is this data?”
Spotter analyzes the dataset, organizing it into high-level categories - like “Analyzing customer churn,” “Understanding customer demographics,” and “Analyzing customer support interactions.”
Within each category, it maps out columns so the business user can understand which data is most relevant and what each column is called.
Spotter also generates example questions tailored to different business functions, jumpstarting analysis
Finally, Spotter returns structured analytical frameworks to guide non-technical exploration, providing clear roadmaps to help you extract meaningful insights and drive business outcomes.
In seconds, Spotter can create a guide for working with this worksheet, highlighting both its structure (columns) and potential applications (questions) in a way that makes the data more accessible and actionable for further analysis. You don’t have to go to an analyst or kick off a BI project; Spotter turns a 6-week BI project into a one-minute discovery session.
Level 3: Complex compute questions
Once you have gotten a sense of your data, you likely want to move into some more specific questions about it. One of the complexities of real-life business questions is that the information required to do the analysis or calculation doesn’t always exist as a simple database column.
When this is the case, Spotter calculates this metric by generating a formula on the fly. The complexity escalates further when the question requires adding additional analytical concepts like a cohort analysis, grouping, and more. This requires multiple layers of computational intelligence to transform raw data into meaningful business insights – which no other tool on the market can do.
In this example, we’re asking, “What is our customer lifetime value by state?
Level 4: Uncover the “Why” with Automated Change Analysis
For business leaders, knowing what happened is just the starting point. Understanding the root cause of the why is critical for strategic decision-making, yet traditionally requires time-consuming analysis from specialized teams. With Change Analysis, Spotter performs multi-dimensional analysis to explain changes in key business metrics instantly.
When asked a “why” question, Spotter follows the same logical process that your favorite analyst would follow. Continuing with the churn example from above, imagine you are looking at churn per month. You see an increase in the chart and ask, “Why is churn increasing in March?”
First, Spotter determines the different dimensions across which they could investigate the question (for example: date, age, churn reason).
Spotter then performs temporal analysis to identify precisely when changes occurred, analyzes what demographics and segments contributed most to churn, and conducts root cause analysis to determine underlying market or operational factors. Each analysis is returned through a data visualization.
But it doesn’t stop there. Spotter also summarizes key findings from each dimension in natural language that reflects your business.
In this way, Spotter can become a dedicated analyst for every decision-maker on your team. When a dedicated agent is performing multi-step, multi-dimensional analysis in seconds, every facet of your business can confidently make decisions in real-time.
Level 5: Partner with your AI Research Assistant through Deep Analysis
When doing analysis, sometimes you know exactly the series of questions you want to ask. Other times, you want to partner with an analyst to dive deep into a broad or topical question. You may not know exactly how to break down the question or even what to investigate—you just know the challenge you are trying to solve.
That’s where Spotter’s Deep Analysis comes in. Spotter helps you tackle broad business questions by breaking them down and then rebuilding with the complete picture. Here’s what that looks like within our churn example: “How do I reduce churn?” This is a big question for any business, and it could take days or even weeks to analyze. But with Spotter as an enterprise-wide agent, any business leader can independently uncover answers in seconds.
Here’s how it works:
First, Spotter breaks the concept into specific questions.
Spotter then analyzes each of these questions, providing a text summary of the analysis and a best-fit visualization—all with no additional human intervention.
Finally, Spotter summarizes all of the findings for instant decision intelligence. It even provides additional questions you may not have thought to ask.
From there, you can keep asking questions and iterating with Spotter, just like you would with your go-to analyst.
Trust Layer: Keeping the human in the loop
This kind of rapid, multi-dimensional analysis is powerful—but only if your team can trust the insights they’re receiving. As the pioneer of search-driven analytics and natural language query (NLQ), ThoughtSpot is uniquely positioned to help you deliver AI with the transparency and accuracy your business requires.
While foundational models like GPT are trained for natural language, they can’t accommodate every complexity of real-world data on their own—think: business context, analytics expressibility, massive and messy datasets. That’s where ThoughtSpot’s architecture comes in. Our Agentic Analytics Platform helps you prepare data for AI, enhance AI’s analytical performance, and provide essential human oversight so you can deliver AI analytics without compromising ease, accuracy, or trust.
Data preparation: With our in-platform Analyst Studio, you can ensure schemas and underlying table relationships are optimized for AI consumption, layering on synonyms for accurate intent recognition.
Analytics architecture: ThoughtSpot’s AI-native architecture and advanced semantic model augment LLM reasoning, using search tokens to improve text-to-SQL accuracy and provide greater transparency.
Human-in-the-loop architecture: With search tokens, every user can verify results in natural language, while data teams can coach ThoughtSpot to capture company- or team-specific vocabulary for consistently accurate AI responses.
In this way, Spotter ensures everyone can see and understand the context behind the analysis, so they can confidently take action and drive positive outcomes across the business.
Transform insights into actions with agentic analytics
Spotter represents a paradigm shift in business intelligence for large enterprises. By putting the power of advanced data science directly into the hands of decision-makers, we're eliminating the traditional barriers between questions and answers and actions. And now, we’re inviting you to experience this power of agentic analytics for yourself—request a demo today.
Spotter is just one part of ThoughtSpot’s Agentic Analytics Platform. See how we’re reimagining BI with AI-first intelligence—and unlock our complete vision of the autonomous enterprise in our latest blog.
