Matillion

Use it when

• Pipelines should be built visually by a mixed-skill team rather than maintained as code.
• The destination is Snowflake, Databricks, Redshift, or BigQuery, and transformations should compile to SQL pushed down to it.
• An enterprise procurement wants one vendor covering load, transform, and orchestration.
• Sources include databases needing CDC, run through an agent in your own cloud.

Think twice when

• Warehouse compute is the hidden line item: every transformation executes on the warehouse, billed by the cloud provider on top of Matillion's credits, and can be the larger of the two bills.
• The destination is not one of those four cloud warehouses; nothing else is a target.
• You want self-serve, code-first workflows; dbt with Airbyte or dlt fits that shape better.
• Budget favors flat or open-source pricing over credit consumption.

How it runs

A managed SaaS with hybrid agents for CDC and private data: orchestration and transformation pipelines are drawn on a canvas and compiled to warehouse SQL, so the warehouse computes and Matillion coordinates. Schedules run in the platform, and credits meter task hours and additional developers.