ClickHouse expands into CDC and ETL
ClickHouse
ClickHouse is turning ingestion into a wedge for selling more database spend, not just a convenience feature. Once a team can point Postgres, MySQL, Kafka, or S3 at ClickHouse with managed connectors, ClickHouse stops being only the place queries run and becomes part of the system that moves production data into analytics. That lets it charge for the flow of data, shorten setup time, and make third party CDC and ETL tools easier to skip.
-
PeerDB gave ClickHouse a fast path into CDC. The July 30, 2024 acquisition brought Postgres replication tech in house, and by late 2025 that engine was fully integrated into ClickPipes. In practice, that means a customer can mirror operational Postgres data into ClickHouse without stitching together a separate replication stack.
-
This changes who ClickHouse competes with. Instead of only fighting Snowflake, BigQuery, or Firebolt at query time, it also starts taking budget from Fivetran and Airbyte, which are paid to move data between systems. Connector revenue matters because ingestion can be metered by data volume and compute, creating a second line item beyond storage and queries.
-
The product effect is tighter workflow control. ClickHouse can shape incoming data into formats that query well inside its own engine, and it can remove the brittle handoff where a separate CDC tool breaks or lags. That is especially valuable for teams syncing Postgres into analytics, where operating one pipeline instead of several cuts time to usable dashboards.
The next step is a more bundled data stack where ClickHouse sells database, ingestion, and increasingly the operational database around it as one system. If that bundle keeps getting easier to turn on, more teams will choose ClickHouse earlier in their architecture and leave dedicated CDC and ETL vendors to serve the harder edge cases, broader connector catalogs, and multi destination deployments.