Move your first pipeline in 15 minutes

The steps

1. Inventory2 minList the connections you want to move and, for each, the source, destination, schedule and which streams are synced. Start with one low-risk pipeline.
2. Create a free account1 minSign up with email or Google. You get a workspace with 1,500 credits a month and no card required.
3. Add the source3 minPick the same connector you use today and enter the same settings. Pipeloom ships Airbyte's connector catalogue, so the fields match. If you are coming from Fivetran, use the nearest connector; most major SaaS sources and databases are covered.
4. Add the destination3 minSame idea. Create a destination for the same warehouse, ideally a separate schema for the trial (for example analytics_pipeloom_test) so the two runs do not collide.
5. Connect them and choose streams2 minDraw the edge on the canvas. Select the same streams and sync modes. Or ask the Copilot to set it up from a description.
6. Run and compare4 minRun the first sync and compare row counts and a few values against the old pipeline's tables. A first run is a full load because sync state does not carry over between platforms.
7. Cut overafter your checksWhen the numbers match, point downstream models and dashboards at the new schema, set the schedule, and pause or delete the old connection. Keep the old one paused for a week as a fallback.

Good to know

• There is no one-click import. You recreate each connection, which is quick because the connector settings are the same.
• Sync state does not transfer. Incremental cursors start fresh, so the first run is a full load. For large tables, plan it off-peak.
• Secrets are re-entered. Credentials are never exported, and Pipeloom stores them in its own secret store.
• Coming from Fivetran? Table and column names and types can differ from Fivetran's schemas, so test downstream models against the new schema before you cut over.

Stuck? Email support@getpipeloom.com and we will help with your first migration.