Understanding how data flows through SQL transformations is critical for debugging, auditing, and refactoring pipelines. FlowScope, part of the PondPilot family, gives you column-level SQL lineage visualization.
What SQL Lineage Shows You
Given a SQL query (or chain of queries), lineage visualization answers:
- Where does each output column come from?
- Which source tables and columns feed into this transformation?
- How does data flow through CTEs, subqueries, and JOINs?
This is essential for impact analysis (“if I change this column, what breaks?”) and regulatory compliance (“prove how this metric is calculated”).
FlowScope: Column-Level Lineage
FlowScope is a Rust-based SQL lineage engine compiled to WebAssembly. It parses SQL statements and produces column-level lineage graphs.
Multi-dialect support: FlowScope handles different SQL dialects — not just DuckDB, but the SQL variants you actually use in production.
Column-level precision: Table-level lineage tells you which tables are involved. Column-level lineage tells you exactly which columns map to which — far more useful for debugging and auditing.
Client-side execution: Like everything in the PondPilot family, FlowScope runs in your browser. Your SQL statements aren’t sent to any server.
Use Cases
Data pipeline debugging: Paste a complex transformation query and see exactly how columns map from source to target.
Compliance and auditing: Regulators want to know how metrics are derived. Lineage visualization provides that proof.
Refactoring: Before changing a column name or removing a table, see what downstream transformations depend on it.
Onboarding: Help new team members understand complex SQL pipelines visually instead of reading hundreds of lines of SQL.
Try FlowScope
Visit FlowScope to paste SQL and see lineage visualized instantly. Open source, free, no signup.