Documentation

Reach out to us at [email protected] to get started.

Schemantic connects with service account credentials for analysis, and the scans and statistical calculations that read your table values run inside your cloud environment. Structural metadata and statistics are stored outside it in a tenant-isolated data plane so the web interface can render; results that are themselves table values may be stored in your environment or by Schemantic.

Operational records — logs, telemetry, traces, diagnostics, cached responses, and similar records — are kept outside your environment to run and support the product. Cross-cloud and AI features change what leaves, and only if you turn them on.

For the full answer, including what the Terms of Service commit us to, see How does Schemantic securely process your data?

When your organization enables AI functionality, such as Chat With Your Data, MCP, or AI-generated analysis, Schemantic sends the AI provider you select the inputs disclosed for that feature. These inputs may include structural metadata, aggregate statistics, descriptive statistics containing table values, and your prompts.

When you authorize an external AI application to connect to Schemantic’s MCP server, that application and its model provider receive your requests and the information returned by enabled tools. Depending on the tools and permissions you grant, this can include metadata, statistics, and table values such as minimums, modes, top-value lists, or query results.

These providers and applications may receive table values, and depending on which AI service you choose, they may retain them. Their processing, retention, and any use for model training are governed by your agreements and settings with that provider. Authorize only the providers, applications, and data access your organization permits.

Before enabling these features, we would encourage you to review Data Handling, Cross-Cloud Features, and AI in our Terms of Service.

1. Open the project
2. Join Map
3. Edit Workspaces
4. Edit data sources
5. Select/deselect data
6. Save data sources
7. Join your data

Universal Filter: Applies workspace-wide across all visualizations. Located at top of Join Map and Entity Hub. Supports plain text search and key:value syntax (table:, dataset:, column:) with AND/OR operators. Filters can be saved, applied before entering workspace, and deleted.

Individual visualizations: Also have their own search boxes with the same syntax.

Yes. SOC 2 Type II attested across all five Trust Service Criteria: Security, Availability, Processing Integrity, Confidentiality, and Privacy.

Check the Schemantic Glossary (link in tool).

Schemantic can surface any value the analysis credential your organization provisions can read, including PII. Your organization controls that credential’s scope and which users it authorizes in Schemantic.