Coalesce Catalog

Coalesce Catalog data catalog

Coalesce Catalog harvests metadata with natural language search, a glossary, automated documentation and a domain-organized marketplace.

Compare: data catalog tools

• Data Catalog: Coalesce Catalog (formerly CastorDoc) is an AI-native data catalog with column-level lineage, classification, AI copilot, and governance — built for Snowflake, Databricks, and Microsoft Fabric (docs).

• Automated Metadata Harvesting: Automated metadata harvesting from connected data platforms and BI tools confirmed (docs).

• Search & Discovery: Natural language search across all data assets with AI-assisted discovery confirmed (docs).

• AI-Powered Search: The cited source (docs/category/ai-assistant) is 404; re-source to AI Search (docs).

• Usage & Popularity Statistics: A 5-star popularity score from query counts (tables) and views (dashboards) (docs).

• Asset Certification: Admins can mark tables as Certified (and sync it to Tableau as tags) (docs).

• Catalog Usage Analytics: User-activity, documentation-coverage and AI-assistant monitoring dashboards for admins (docs).

• Business Glossary: Re-source to Knowledge pages (the glossary terms) (docs).

• Metadata Propagation: Column descriptions propagate downstream through lineage when columns pass through unchanged (docs).

• Custom Fields & Templates: Description templates and custom ownership labels exist, but no custom metadata fields or properties are documented (docs).

Coalesce Catalog data lineage

Coalesce Catalog shows column-level lineage from the warehouse to BI tools like Power BI, Looker, Tableau, Sigma and Omni, with lineage from Coalesce Transform and Quality at no setup.

Compare: data lineage tools

• Data Lineage: Column-level lineage confirmed — tracks field-to-field dependencies across transformations and BI tools (docs).

• Table-Level Lineage: Table-level lineage confirmed as part of the full lineage graph (docs).

• Column-Level Lineage: Column-level lineage is a highlighted capability — traces column dependencies from source through transformations (docs).

• End-to-End Lineage: End-to-end lineage from source through transformations to BI layer confirmed (docs).

• Impact Analysis: Impact analysis confirmed — identifies downstream assets affected by upstream changes (docs).

• OpenLineage Support: Airflow lineage through OpenLineage into Marquez, which Catalog reads; no native OpenLineage endpoint (docs).

Coalesce Catalog data stewardship

• Review & Approval Workflows: Read Me Content Review (2026-08-26): non-owners propose, owners and admins approve (docs).

• Governance Campaigns: The Metadata Editor bulk-applies owners, tags and descriptions to all assets matching filters (2026-02-12) (docs).

• Metadata Automation: Re-source to Metadata Automations (2026-07-29, rule-based owner assignment) instead of AI administration (docs).

• Data Stewardship: Ownership and stewardship workflows not explicitly confirmed beyond RBAC (website).

• Critical Data Elements: No CDE registry in the docs; the closest are Certified and pinned assets (docs).

• Governance Scorecards: Only documentation-coverage percentages, no questionnaire-based scoring (docs).

Coalesce Catalog data classification

• Sensitive Data Discovery: Keep, but note it detects PII from metadata patterns only, not data values; re-source (docs).

• Data Policy: "Manage access" only limits what teams see in Catalog and "doesn't replace warehouse object privileges such as Snowflake grants"; the cited RBAC page is about Coalesce Transform (docs).

• Native Policy Enforcement: Catalog access settings don't push grants or masking to warehouses; sync-back covers descriptions and Tableau tags only (docs).

• Data Masking: Catalog reads metadata only and has no masking (docs).

Coalesce Catalog AI features

The AI assistant works in Slack, Teams and Google Chat, and the MCP server (December 2025) uses per-user OAuth and can write descriptions and glossary terms.

• AI Features: Coalesce Copilot (AI SQL assistant + documentation), AI classification, and AI-powered search all confirmed (docs).

• AI Chat Assistant: Rename "Coalesce Copilot" to the Catalog AI Assistant (AI Search, SQL Copilot, Dashboard Q&A; in Slack, Teams, Google Chat); the old source is 404 (docs).

• AI Documentation Generation: Automated AI documentation generation for tables and columns confirmed (docs).

• Agentic Metadata Enrichment: AI descriptions are user-triggered or come through external agents via MCP; Automations are rule-based owner assignment (docs).

Coalesce Catalog data compliance

• Audit Trail: Re-source from assets/tables to the History tab (2025-12-04), which logs who changed descriptions, owners, tags, certification and PII (docs).

• Data Compliance: Compliance features not explicitly confirmed beyond audit trails (website).

• Compliance Reporting: Only PII column reports and coverage dashboards, no compliance evidence reports (docs).

• Regulatory Framework Mapping: No regulation mapping in the docs (docs).

• Data Activity Monitoring: Query logs feed popularity only; no access monitoring or alerting (docs).

• ROPA: Out of scope for a metadata catalog (docs).

• Data Retention Policies: No retention management (docs).

• Data Subject Rights: No DSR workflows (docs).

Coalesce Catalog collaboration

• Collaboration: Slack and Microsoft Teams integration for notifications and team-based data discovery confirmed (docs).

• Alert Notifications: Re-source to the notifications page (Slack, Teams, email; direct and digest) (docs).

• Discussions & Annotations: Every asset has a Comments section with @mentions, and Slack threads can be saved to it (docs).

• Ticketing Integrations: No Jira or ServiceNow integration; issue reports go to owners through comments, Slack or email (docs).

Coalesce Catalog data marketplace

• Self-Service Access Requests: Users can request table access from the owner in Catalog; this notifies the owner but does not provision access (docs).

• Data Products: Assets published as data products to the Marketplace, by domain (Jul 2025), with API endpoints (2026-03-09) (docs).

• Data Mesh Support: Domain-organized Marketplace with publishing criteria, domain tags and owner labels (docs).

• Data Marketplace: Not confirmed (website).

• Subscription Management: Notifications go to asset owners; no consumer subscriptions to data products are documented (docs).

Coalesce Catalog data quality

Coalesce Catalog shows data quality test results but doesn't run checks itself; that's Coalesce Quality.

• Data Quality: Data quality monitoring not confirmed as a native Coalesce Catalog feature (website).

• SQL Table Tests: "Catalog does not run tests itself"; it shows results from dbt, Soda, Monte Carlo, Sifflet or Coalesce (docs).

• Data Health Scores: The quality dashboard shows pass/alert/warning per table, not a composite score (docs).

• Anomaly Detection: Not in Catalog; anomaly monitors are in the separate Coalesce Quality product (docs).

• Data Profiling: Catalog reads metadata only, so no data profiling (docs).

• Data Contracts: No contract definitions in the docs (docs).

Coalesce Catalog workflows

• Workflows: API and collaboration integrations (Slack, Teams, Notion) confirmed (docs).

• REST / GraphQL API: API access confirmed for catalog programmatic integration (docs).

• MCP Server: Hosted Catalog MCP (2025-12-08) with OAuth, search, lineage and description-write tools (docs).

• dbt Integration: Ingests dbt manifests and test results, and syncs descriptions back to dbt YAML (docs).

• Browser Extension: Chrome and Edge extension shows descriptions, owners and popularity in Looker, Tableau, Power BI, Sigma and more (docs).

• Webhooks: No outbound webhooks documented; integrations use the GraphQL API (docs).

• Python SDK: No Python API client; castor-extractor is a Python CLI for metadata extraction only (docs).

Coalesce Catalog data context layer

• Semantic Models: Marketing mentions a "metadata-powered semantic layer", but the docs show no semantic model generation (docs).

• Knowledge Graph: No graph model or graph queries are documented; Knowledge Map is a diagram of glossary pages (docs).