Cross-warehouse, not Snowflake-only
BoltPipeline runs across Snowflake + Postgres today, with more warehouses on the roadmap. Horizon is Snowflake-only by design.
BoltPipeline governs the pipeline that produces data; Horizon governs the data at rest inside Snowflake. Different layers — they pair.
BoltPipeline vs Snowflake Horizon on the capabilities that drive the buying decision.
| Capability | BoltPipeline | Snowflake Horizon |
|---|---|---|
| Cross-warehouse (Postgres + Snowflake + roadmap) | Yes | — |
| Pre-deploy certification gate | Yes | — |
| AST column lineage across warehouses | Yes | Partial |
| Snowflake-native access policies | — | Yes |
| Tag-based masking inside Snowflake | — | Yes |
| Data-at-rest governance | — | Yes |
BoltPipeline runs across Snowflake + Postgres today, with more warehouses on the roadmap. Horizon is Snowflake-only by design.
BoltPipeline blocks bad pipelines before they ship. Horizon enforces who can read what after the data is already there. Different governance posture.
AST column lineage works across Snowflake + Postgres + roadmap warehouses. Horizon lineage stops at the Snowflake boundary.
No — they're different layers. Horizon governs data-at-rest inside Snowflake (tags, masking, access policies). BoltPipeline governs the pipeline that produces and updates that data. Run both.
Horizon lineage is Snowflake-native and only covers Snowflake objects. BoltPipeline lineage is AST-derived and works across Snowflake + Postgres + roadmap warehouses, with the derivation expression on each column edge.
Yes — this is the recommended pattern for Snowflake shops. BoltPipeline gates the pipeline at deploy time; Horizon enforces access policies on the resulting data at query time.
Try BoltPipeline against your live database — your data never leaves your environment.