Looker - Lineage
This article outlines lineage coverage, configuration requirements, supported components, transformation support, supported use cases, partial or limited coverage, unsupported scenarios, and the current functional status of lineage extraction in Looker.
The Looker connector provides end-to-end lineage from underlying database tables to Looker Looks and Dashboards. It supports lineage across SQL-based reports and dashboards, including column-level lineage, to provide visibility into data flow and impact analysis. The connector supports static SQL queries, derived tables, merge queries, and supported SQL transformations.
Lineage Configuration Requirements
Successful lineage extraction depends on valid Looker API access, appropriate permissions, available database metadata, accessible SQL definitions, and correct SQL dialect configuration. The following requirements apply to Looker lineage extraction.
Configuration Requirements Table
Authentication
Valid Looker API credentials and connection setup must be available for metadata extraction.
Access
Read access to Looks, Dashboards, and the underlying database must be available.
Database Metadata
Database metadata must be crawled before lineage is built so that source tables and columns can be resolved.
SQL Availability
SQL queries must be available for lineage generation.
SQL Dialects
The appropriate SQL dialect must be configured to support SQL parsing.
SourceCode
SQL definitions must be available in SourceCode for lineage generation.
Missing API access, insufficient permissions, unavailable database metadata, inaccessible SQL definitions, or incorrect SQL dialect configuration may result in incomplete lineage generation.
Lineage Components
The following table identifies the Looker components and lineage features handled by the connector.
Looks
✅
Dashboards
✅
Dashboard Tiles
✅
Merge Queries
✅
Static SQL Queries
✅
Derived Tables
✅
LookML Views / Explores
⚠️
Calculated Fields (SQL-based)
⚠️
Window Functions
✅
Dynamic SQL
❌
Parameterized Queries
❌
Table Calculations (Frontend)
❌
Column Creation Support
This section outlines lineage support for columns created through direct references, aliases, aggregations, CASE statements, window functions, expressions, and LookML SQL-based dimensions.
Column Creation from Lineage
Direct Column References
✅
Column Aliases
✅
Aggregations
✅
CASE Statements
✅
Window Functions
✅
Expressions (Arithmetic / String)
✅
Dot Notation Handling
✅
Table Calculations (Frontend)
❌
LookML Custom Dimensions (SQL-based)
⚠️
Supported Use Cases
The Looker connector supports lineage extraction across SQL-based Looks, Dashboards, Dashboard Tiles, and crawled database objects.
Supported Lineage Scenarios
SQL-based Looks
Lineage from underlying database objects to Looks through available SQL definitions.
Dashboards with Multiple Tiles
Lineage across Dashboards and their associated Tiles.
Column-level Lineage from SQL
Column-level lineage from source database columns through SQL-based Looker content.
Reports Using Crawled DB Objects
Lineage for reports that reference database tables and columns available in the crawled catalog.
Multi-dialect Environments
SQL lineage across supported database dialects when the corresponding SQL dialect configuration is available.
Static SQL Query Lineage
Lineage from source tables and columns through statically available SQL queries.
Derived Table Lineage
Lineage involving derived tables represented in the available SQL definitions.
Merge Query Lineage
Lineage across Merge Queries where source attribution can be resolved from the available SQL.
Window Function Lineage
Column-level lineage for supported SQL transformations that use window functions.
SQL-based Calculated Field Lineage
Lineage for calculated fields when the calculation is represented in the available SQL.
Partial or Limited Coverage
Certain Looker scenarios provide partial lineage coverage because the required SQL may not be available, LookML definitions may not be fully parsed, or source attribution may become complex across multiple query components.
Scenarios
LookML-based Reports Without SQL
LookML-based reports require SQL extraction, and LookML definitions are not fully parsed for lineage.
Ensure SQL definitions are available for the affected reports and validate LookML-specific dependencies manually where required.
Calculated Fields
Calculated fields are supported only when the calculation is represented in the available SQL.
Validate calculated-field dependencies separately when the calculation is not represented in SQL.
Derived / Temporary Tables
Derived or temporary tables may appear as generic objects when the corresponding objects are not available in the catalog.
Ensure relevant source metadata is crawled and validate derived or temporary table dependencies where required.
Merge Queries
Complex attribution across multiple sources may limit complete lineage resolution.
Validate source attribution manually for complex Merge Queries.
LookML Custom Dimensions
SQL-based custom dimensions provide partial lineage coverage when the required SQL is available.
Ensure SQL definitions are available and validate complex LookML dimensions manually.
LookML-based content without accessible SQL, calculated fields that are not represented in SQL, uncataloged derived or temporary tables, and complex Merge Queries may reduce lineage completeness and column-level accuracy.
Unsupported Scenarios
The connector does not support lineage extraction for runtime-generated SQL, unresolved query parameters, or frontend-only calculations that do not form part of the SQL lineage.
Unsupported Lineage
Dynamic SQL
Runtime-generated SQL and runtime parameter behavior are not resolved during lineage extraction.
Use stable SQL definitions where possible and validate dynamic dependencies manually.
Parameterized Queries
SQL placeholders and parameter values are not interpreted for lineage generation.
Validate parameter-dependent source relationships separately where required.
Table Calculations
Frontend-only Table Calculations are not part of SQL lineage and are not captured as source-to-target SQL relationships.
Validate frontend calculation dependencies separately where required.
Frontend-only Calculations
Calculations that do not appear in the extracted SQL are not available for SQL-based lineage generation.
Ensure applicable calculations are represented in SQL where lineage coverage is required.
Current Functional Status
This section outlines the current lineage capabilities and coverage of the Looker connector.
Overall Coverage
Production-ready coverage across SQL-based Looks, Dashboards, Dashboard Tiles, static SQL queries, derived tables, and supported database sources.
Lineage Depth
Table-level and column-level lineage from underlying database objects through SQL-based Looker content.
Supported Inputs
Looks, Dashboards, Dashboard Tiles, Merge Queries, Static SQL Queries, Derived Tables, supported SQL-based calculated fields, and crawled database objects.
Functional Scope
End-to-end lineage from underlying database tables and columns through SQL queries to Looker Looks and Dashboards.
Transformation Support
Direct column references, aliases, aggregations, CASE statements, window functions, arithmetic and string expressions, and dot notation handling are supported.
Limitation Areas
LookML-based reports, LookML custom dimensions, SQL-based calculated fields, derived or temporary tables, and complex Merge Queries have partial coverage.
Unsupported Areas
Dynamic SQL, parameterized queries, and frontend-only Table Calculations are not supported for lineage extraction.
Resulting Output
Robust lineage for structured SQL-based Looker environments, including column-level tracking, with partial coverage for LookML-based constructs and unsupported coverage for dynamic and parameterized query scenarios.
The Looker connector provides production-ready lineage coverage for SQL-based Looks and Dashboards, including column-level lineage across supported database sources. Direct column references, aliases, aggregations, CASE statements, window functions, expressions, and dot notation are supported. LookML-based constructs, SQL-based calculated fields, derived or temporary tables, and complex Merge Queries have partial coverage, while dynamic SQL, parameterized queries, and frontend-only Table Calculations are not supported. Manual validation may be required for advanced or SQL-independent scenarios.
Copyright © 2026, OvalEdge LLC, Peachtree Corners, GA, USA.
Last updated
Was this helpful?

