Fivetran

Automated data movement

Managed ELT platform with 300+ connectors for automated data pipeline maintenance.

Our take

Fivetran solves the worst part of data engineering: building and maintaining connectors. Every data source (Salesforce, Postgres, Shopify, Google Ads) has schema quirks, API rate limits, and version changes that break pipelines. Fivetran maintains 300+ connectors so data engineers don't have to — data flows from source to warehouse automatically, with automatic schema migration when source schemas change. The pricing is based on Monthly Active Rows (rows synced), which scales with data volume rather than seat count. The cost can be significant at enterprise data volumes — some teams report $30,000-$100,000/year — and the MAR model is opaque. Alternatives like Airbyte (open-source, self-hosted) and Stitch (Talend, cheaper) handle common connectors at lower cost. Fivetran's value proposition is reliability and maintenance: the connectors just work, and Fivetran engineers fix them when source APIs change. dbt relationship: they're the standard pipeline pair — Fivetran (extract + load) + dbt (transform).

Product roadmap

Fivetran is building out its data movement platform beyond batch ETL — real-time CDC (Change Data Capture), reverse ETL (pushing warehouse data back to operational tools), and data lake support (writing directly to S3/GCS). The HVR acquisition brought high-volume replication capabilities for large database migrations.