Skippr Docs

Appearance

How to Use DynamoDB Streams for Warehouse CDC ​

April 2026

DynamoDB Streams can drive warehouse CDC cleanly if you make row identity and destination behavior explicit.

Short Answer ​

Use DynamoDB Streams for warehouse CDC when you want ongoing change capture instead of periodic exports. The important part is not just turning the stream on. It is making row identity and destination behavior explicit.

Skippr supports DynamoDB CDC through Streams, then reconciles those changes into warehouse tables with the right final-state semantics.

Why Teams Struggle with This ​

Teams often get the stream running and then stall on the warehouse contract. The events move, but nobody is sure how updates, deletes, and row identity should behave downstream.

  • CDC gets enabled without clear business keys.
  • The destination table shape is left implicit.
  • Deletes are captured but not handled well downstream.
  • The warehouse gets events, not a table contract people trust.

How Skippr Handles It ​

Skippr keeps the setup narrow. Enable CDC on the source, define the CDC block, declare business keys, and let the warehouse destination handle the merge behavior.

That gives the team a more useful starting point than shipping a stream into a pile of downstream clean-up jobs.

  • DynamoDB Streams support for CDC.
  • Business-key-driven reconciliation.
  • Destination-aware final-state behavior.
  • Generated dbt project on top of the landed warehouse tables.

What the First Useful Version Looks Like ​

The hard part is not turning on the stream.

The hard part is landing a warehouse table that still looks sane after real traffic.