Coming soon

The following features are coming in the next release of ThoughtSpot Cloud.

October 2026 26.10.0.cl

The spreadsheet interface puts a familiar, editable grid directly on your live data. Save your work as an Answer that stays searchable, shareable, and governed like anything else in ThoughtSpot.

For more information, see Spreadsheet.

Spotter now features a browsable catalog with more than 100 third-party connectors, making it easier than ever to discover and add third-party data integrations. Every user can browse, search, and explore available connectors, including one-click options like Slack and Web Search, plus connectors for services like Microsoft Teams that guide you through setup. Your administrator or users with the CanManageSpotter privilege can add and configure connectors for your organization. If you don’t see a specific MCP server in the catalog, any MCP-compliant server can still be added manually. The Spotter marketplace connectors are available to all users in Spotter by selecting Settings > Connectors. This feature will be rolled out in the first weekly patch after the 26.10.0.cl release.

The new charts library expands to add Pareto, Waterfall, and Heatmap charts, bringing enhanced analytical flexibility and modernized styling across search, Spotter, and Liveboards. This release introduces support for multi-panel trellis and faceting modes, new multivariate axis configuration logic, and expanded customization options — including independent color controls for increment and decrement bars on Waterfall charts, refined gradient and label positioning on Heatmaps, and enhanced axis formatting on Pareto charts.

Liveboard error management is now generally available (GA). When a data model column that a Liveboard depends on is deleted, ThoughtSpot now detects the impact and shows you exactly what’s broken instead of displaying a generic error or silently failing.

The centralized filter window is now generally available (GA).

For more information, see Centralized filter window.

Liveboard creators can now scope filters and parameters to a specific tab or group, instead of applying them globally across the entire Liveboard. This makes it easier to build focused, information-dense Liveboards where consumers can analyze a specific area without inadvertently affecting unrelated visualizations. To enable this feature, contact your administrator.

When a Liveboard with tab or group filters is imported into a cluster where this feature is disabled, the scoped filters are removed on import.

For more information, see Liveboard filters.

The enhanced behavior of Liveboards built with column security rules is now generally available (GA). If users restricted by column security access the Liveboard, they can see and interact with the Liveboard filters and visualizations, but cannot edit the Liveboard as a whole. If such inaccessible columns are used as filters in the Liveboard, the Liveboard will no longer be blocked. Schedules now honor column security rules, and each recipient receives their own report based on their column security rule assignments.

For more information, see Column security rules on Liveboards.

The Spotter analyst builder is now generally available (GA). In this chat-based interface, you can describe the analyst you would like to use, and Spotter translates your requests into instructions for the agent. Spotter Analysts are also GA in this release.

For more information, see Spotter analysts.

We’ve added more granular controls for sharing of Spotter connectors in ThoughtSpot. Previously, if an administrator enabled a Spotter connector like Slack, it was shared across all Orgs. Now, if you add a connector in the primary Org, you can choose whether to share it with certain users and groups, and it will be hidden by default from all other users and groups. You can only add, share, or edit connectors from the primary Org.

ThoughtSpot writeback tables are now generally available (GA).

For more information, see Writeback.

Muze Studio now includes an AI-powered chat assistant that lets you build and refine custom charts using natural language. Simply describe the visualization you want, or ask follow-up questions to iterate on it, and the AI generates the configuration for you directly inside the Muze Studio editor. A built-in "Show work" timeline lets you trace each step the AI took, giving you full visibility into how your chart was constructed. Muze Studio AI is available in Beta on ThoughtSpot Cloud and can be enabled by your administrator.

Publishing from object lists is now generally available (GA). You can now publish, unpublish, and manage the publishing state of Answers, Liveboards, and Models directly from the list page view without having to open each object individually.

The Publish, Unpublish, and Manage publishing options are available in the row-level action menu on the Answers, Liveboards, and Data Workspace list pages. The publish and unpublish options allow you to select target Orgs, handle dependents, and confirm the action.

Publishing is a multi-Org feature. These actions are only visible in multi-Org environments where object publishing is enabled. You must be in the primary Org and be a cluster administrator to see the publish and unpublish actions.

For Models specifically, the publish action is not available for objects authored by a superuser or system user.

You can now set query timeout in the Advanced Config setting in Connections, where you can define how many seconds a query should run for before it automatically times out.

ThoughtSpot administrators can now impersonate a specific user to troubleshoot issues directly, without requiring screen shares or shadow accounts. To protect user privacy, each user must explicitly enable the Act as setting in their profile before an administrator can initiate a session. Sessions are limited to 60 minutes, and the administrator can end the session at any time using the Exit session button. The target user receives email notifications when a session starts and ends. All actions taken during the session are captured in audit logs.

To enable this feature, contact ThoughtSpot support.

Databricks Metric Views integration is now available in ThoughtSpot. This integration allows you to connect directly to Metric Views defined in Databricks and use them to automatically create Models in ThoughtSpot. You can import Metric Views, including metrics, formulas, and column metadata, and immediately use them for building Liveboards, running queries, analytics, and Spotter. Users with Can manage data privileges can manage and resync Metric Views as needed to reflect changes made in Databricks. To enable this feature, contact your administrator.

ThoughtSpot now uses AI to translate external semantic definitions into ThoughtSpot formulas and Models during import, delivering higher accuracy than the previous rule-based translator, especially for complex SQL expressions that the deterministic parser could not handle. This feature is supported for Snowflake Semantic Views, Databricks Metric Views, and dbt MetricFlow imports. For Databricks Metric Views, you can provide the Metric View definition either by uploading a YAML file or by connecting directly to Databricks to retrieve the definition at import time. To enable this feature, contact ThoughtSpot support.

When importing a dbt MetricFlow semantic model from Snowflake or Databricks, ThoughtSpot now uses AI to translate metric measures and dimension expressions into ThoughtSpot formulas.

The new Data workspace page is now generally available (GA). It now uses a modernized object table, making it easier to find, filter, and manage your data objects in one place.

ThoughtSpot now enforces row-level security (RLS) on saved filter values in Answers and Liveboards. When a filter chip displays values from a column marked as Secure Suggestions in the data model, ThoughtSpot checks whether the current user has RLS access to those values before displaying them.

When filter values are cleared due to RLS enforcement, the user still sees data, but only the rows their RLS policy permits. An empty filter chip on a secured column does not mean no data; it means the saved filter values were not applicable to this user and the RLS policy is the active constraint.

To enable this feature, contact ThoughtSpot support.

You can now define object IDs for variables in TML or the API. Object IDs now support multi-tenant objects. Use object IDs in tandem with GUIDs that identify objects in TML. You can use object IDs to identify objects that you export in TML and import to other Orgs, removing the need to completely recreate objects between Orgs.

We have upgraded the Amazon Athena driver to version 3.8.0.

For new features and enhancements introduced in this release of ThoughtSpot Embedded, see ThoughtSpot Developer Documentation.