Skip to main content
The Fivetran importer reads connector and sync metadata so you can recreate ingestion intent as Planasonix pipelines. You still map credentials to connections and validate transforms on the canvas.
Start with Import and convert for the shared wizard flow, batch import, and validation discipline.

Discovery API integration

When Planasonix supports the Fivetran REST API, you authorize a read-scoped token and let the wizard discover connectors, destinations, and sync schedules.
1

Create API credentials

In Fivetran, create an API key with permissions to list connectors and read configuration metadata. Store the key in Planasonix’s secret manager when prompted.
2

Run discovery

Enter account identifiers the API requires. Planasonix pulls connector types, source schemas, table include/exclude rules, and column settings where exposed.
3

Select scope

Choose specific connectors or tags to convert. Large accounts should migrate by domain (for example, finance vs product analytics) to keep review manageable.
Export a connector inventory spreadsheet before migration. Use it to track owner, SLA, and whether the connector is still needed—API discovery can surface deprecated connectors you should retire instead of porting.

Connector mapping

Fivetran connectors map to Planasonix source nodes, staging patterns, and optional orchestration that mirrors sync frequency.
Database connectors become ** JDBC/API source** configurations plus incremental or full-table strategies inferred from Fivetran sync modes (for example, history mode vs standard incremental). Verify primary keys and cursor columns after import; Fivetran sometimes hides complexity behind its own change-data layer.

What transfers vs what needs manual work

  • Source type and high-level connection parameters (host, database, API object selection)
  • Table and column inclusion lists when exposed by the API
  • Schedule cadence as a starting point for Planasonix schedules or triggers
  • Transforms beyond Fivetran’s column blocking/hashing—Planasonix models these as explicit transform nodes
  • Webhook or event-based loads if you replace sync schedules with cloud triggers
  • Secrets and network paths (PrivateLink, SSH tunnels) that cannot be exported in clear text
  • Fivetran dbt transformations if you used them—treat as dbt logic to merge separately or reimplement on the canvas
Fivetran historical sync behavior and deleted record handling are product-specific. Compare row counts and slowly changing dimensions in a sandbox warehouse before you cut over production consumers.

After conversion

Pipeline canvas

Refine graphs, preview data, and attach tests.

Variables

Externalize environment-specific names and paths.