Data & Analytics · ThoughtSpot Spotter MCP + Claude
ThoughtSpot Spotter MCP with Claude: Governed Answers From Your Business Data
The short answer
Claude works with ThoughtSpot Spotter through the Model Context Protocol (MCP), an open standard that lets Claude securely use other software. Once connected, Claude can query your governed data, ask for natural-language answers, run root-cause analyses and export visualisations, with permissions your organisation controls.
What is ThoughtSpot Spotter MCP?
ThoughtSpot Spotter is an enterprise analytics platform built around natural-language search and governed insights. Business users type questions in everyday language and receive answers drawn from approved data sources. Data teams, finance leaders, sales managers and operations heads use it to find out what is happening in the business and why. Its focus is on trusted, explainable analytics that can be used at scale across an organisation.
How Claude connects to ThoughtSpot Spotter MCP through MCP
The Model Context Protocol (MCP) is an open standard that works like a universal adapter between Claude and other software. It lets Claude use a tool securely, reading information and taking actions, with permissions the organisation sets. ThoughtSpot Spotter supports connection through its MCP server or a direct API, and sign-in uses DCR OAuth, a standard sign-in method that the organisation controls.
Once connected, Claude can query governed data in ThoughtSpot, generate natural-language answers, run root-cause analyses to explain why a figure changed, and export visualisations on request. Claude works only with the data and permissions the connected account is allowed to see, so answers stay within existing governance rules.
Permissions and approvals keep people in charge. Access is limited to what each role needs, and consequential actions, such as sharing an export outside the company, can require a person to approve them first. Audit logs record activity, so teams can review how each answer was produced.
What you can automate
Weekly Sales Performance Briefings
A sales director asks Claude for the week's pipeline movement by region. Claude queries governed sales data in ThoughtSpot and returns a short, explained summary, highlighting regions that moved most and why. The director reviews the answer before sharing it with the leadership team, saving hours of manual report building each week.
Root-Cause Analysis for Margin Changes
A finance manager notices gross margin dropped in one product line. Claude runs a root-cause analysis in ThoughtSpot to identify the cost, pricing or volume factors behind the change. The explanation gives finance a starting point for review, so decisions rest on evidence rather than guesswork.
Operations Exception Monitoring
An operations head wants early warning of delivery delays. Claude checks governed operational data each morning, asks ThoughtSpot for exceptions outside agreed thresholds, and drafts a list for the team. People decide which exceptions need action, and the list is stored with an audit trail.
Self-Service Answers for Non-Analysts
Staff in marketing or HR often need a quick figure without raising a data request. They ask Claude a plain-language question, and Claude retrieves the governed answer from ThoughtSpot. This reduces queue times for the data team while keeping every figure tied to approved definitions.
Exporting Charts for Board Packs
A chief of staff prepares a monthly board pack. Claude requests the relevant visualisations from ThoughtSpot and exports them in the requested format. The chief of staff checks each chart for accuracy before it is added to the pack, keeping control of the final document.
An end-to-end workflow: Monthly Revenue Review With CRM Context
TriggerThe first working day of each month, a finance analyst asks Claude for the revenue review.
Claude queries governed revenue data in ThoughtSpot Spotter for the previous month, by region and product.
Claude runs a root-cause analysis on any region that missed its target by more than the agreed threshold.
Claude pulls the related account and deal notes from the CRM to add commercial context to the explanation.
Claude drafts a written summary with the key figures, causes and supporting visualisations.
The finance analyst reviews the draft, corrects anything unclear and approves it for distribution.
The approved review is saved to the shared drive and logged in the audit trail.
OutcomeTypically, the monthly review is ready on the first day rather than several days later, and each explanation links figures to the commercial context behind them. Final judgement and approval remain with the finance team.
How we set it up
Start With a Free Automation Audit
We review your current reporting and analytics workflows with you and identify the questions worth automating first. You receive a clear list of options and recommended priorities at no cost.
Confirm Access and Governance
We check your ThoughtSpot setup, data models and permission structure. We agree which data sets Claude may use and who can approve sharing or exports.
Connect Claude Through MCP
We connect Claude to ThoughtSpot Spotter using its MCP server or direct API, with DCR OAuth sign-in. Access is set to least privilege, so each connection sees only what it needs.
Design, Test and Launch
We build the end-to-end workflow, test it against real past questions and reports, and check that answers match known figures. We launch with monitoring and audit logs turned on.
Train Your Team
We train analysts, managers and business users on how to ask good questions, how to read explanations and when to escalate. Ongoing support is available as your usage grows.
Who benefits
- Chief Financial Officer
- Faster, explained answers on margins, costs and forecasts, with a clear audit trail.
- Head of Sales Operations
- Weekly pipeline and performance briefings produced in minutes rather than days.
- Data and Analytics Lead
- Fewer routine data requests, with governed definitions kept intact.
- Operations Director
- Earlier warning of exceptions, with people deciding what happens next.
Security and control
- Claude only accesses the governed data and permissions granted to the connected account.
- Least-privilege access limits each connection to the minimum data it requires.
- People approve consequential actions, including sharing and exporting outside the organisation.
- Audit logs record queries and outputs so every answer can be reviewed.
- Hosting region is chosen to meet your data residency requirements before launch.
For teams in the US, UK, India and the Middle East
Organisations in the United States and the United Kingdom can align the connection with their existing data governance and privacy obligations. Businesses in India must consider the Digital Personal Data Protection Act, 2023, when handling personal data. Organisations in the Middle East, including the UAE and Saudi Arabia, must comply with local data protection laws. We select a hosting region in India, the United States, the United Kingdom or the European Union to meet the requirements that apply to you.
Learn more about Claude and MCP
Anthropic's AI model and assistant.
Secure links between Claude and business tools.
How engineers build and change software with Claude.
Hosted agents that run multi-step work.
The open standard Claude uses to connect to tools.
Frequently asked questions
What does the ThoughtSpot Spotter MCP server do?
The MCP server gives Claude a secure way to work with ThoughtSpot Spotter. It allows Claude to query governed data, request natural-language answers, run root-cause analyses and export visualisations. The server acts within the permissions of the connected account, so it does not give broader access than your organisation has set.
Is connecting ThoughtSpot to Claude secure?
The connection uses DCR OAuth sign-in, a standard method that avoids sharing raw passwords. Access is limited to the data and actions each connection needs. Audit logs record activity, and people approve consequential actions. We design the setup with your security and governance teams before launch.
What can Claude do with ThoughtSpot once connected?
Typically, Claude can ask ThoughtSpot governed data questions in plain language, receive explained answers, run root-cause analyses to understand changes, and export visualisations. Exactly what is possible depends on how your ThoughtSpot environment and permissions are configured.
How long does setup take?
Timelines vary with the number of workflows, data sets and approval steps involved. A focused first workflow is often ready within a few weeks, following the free audit, access checks, testing and training. We confirm a realistic timeline during the audit.
Which ThoughtSpot plan or permissions are needed?
The requirements depend on your ThoughtSpot setup and the MCP or API access your organisation has enabled. In general, the connected account needs appropriate permissions to the data sets you want Claude to use. We review your current licence and permissions during the audit and advise on any changes needed.