Ingestion
Bring your business data together.
Connect your apps and databases to your warehouse so your team can analyze them together. Choose the data you need, set a schedule, and keep your reporting data up to date.
Bring the data you need
Select source objects and fields instead of moving everything by default.
Choose how tables stay current
Use full or incremental syncs where supported, with a load mode suited to your destination.
Check what loaded
See rows ingested, loaded, and failed, with a breakdown for each object.
Choose your data
Bring only the fields you use.
Start with the objects your team needs, then narrow each one to the fields your reports use. Review field types, keys, and available cursor settings before the data reaches your warehouse.
- Choose individual source objects
- Select the fields to include in each object
- Review field types, keys, and cursor settings
Sync and load
Choose how your tables stay current.
Start with historical data, then keep it current with incremental syncs where supported. Control how incoming records update your tables and which source schema changes carry through.
- Full and incremental sync modes where supported
- Append records, merge updates, or reload tables where supported
- Control which source schema changes reach your tables
Sync results
Know what happened after each sync.
See which objects loaded, how many rows reached the warehouse, and which objects failed. Use the Ingest and Load stage results to investigate a missing update.
- Object counts and execution status
- Ingest and Load stage metrics
- Failure and warning email preferences
In practice
Bring CRM data into reporting.
Analyze Salesforce accounts and opportunities alongside the rest of your business data in Snowflake. Use the same warehouse tables across your CRM reports.
- Bring the accounts and opportunities your reports depend on into Snowflake.
- Keep reporting tables current with the sync modes your sources support.
- See the delivery history when someone asks about a missing update.
Available sync modes, incremental behavior, and schema options depend on the selected connectors.
Questions before you start
Can I bring in just the data my reports need?
Yes. Select individual objects and fields for each pipeline. You can review their types, keys, and available cursor settings before syncing.
Can I load historical data, then keep up with new changes?
Choose Historical + Incremental for an initial full load followed by changes where supported. Historical Only and Incremental Only are also available for compatible sources. You can schedule full refreshes for objects without incremental support and periodic full resyncs for the pipeline.
Can I update existing records instead of rebuilding my tables?
Yes. Use Merge to insert new records and update existing ones using a record key. You can also append records or refresh tables with Truncate and Load or Overwrite where the destination supports those modes.
Can I decide which source schema changes reach my warehouse?
Yes. Allow all changes, block changes, or allow column-level changes only. You can also choose whether new tables, new columns, or schema changes trigger a full resync, where supported.
Can I control the names of replicated objects?
Yes. Namespace rules let you mirror source namespaces, use the destination’s configured schema, or apply Fivetran-style naming where supported. Pipelines with an editable schema or prefix also let you set your own namespace.
Can I protect existing tables when a pipeline first syncs?
Yes. Choose how the pipeline handles an existing destination table. For example, Fail if Table Exists lets you stop a naming collision instead of merging into or replacing that table. Available collision options depend on the connectors.
How can I lower warehouse costs for pipeline loads?
Use Optimize for Cost to batch operations where supported. Match the sync schedule to how often your reports need fresh data, and select only the objects and fields your team uses. Available optimization modes depend on the destination.
Can I choose how data is checked and errors are handled?
Yes. Choose validation checks for row counts or both schema and counts where supported. For row-level errors, choose whether to log the error and continue or stop the affected object while other objects continue. Validation is optional; available checks and error controls depend on the connectors.
Can I pause a pipeline or reload its data without starting over?
Yes. Disable a pipeline to stop scheduled runs, then enable it when you are ready. Run Sync Now when you need an update, or use Resync Data to reload the source without relying on its previous incremental position. You can also edit settings and object selections on the existing pipeline.
Can I choose which pipeline updates reach my inbox?
Yes. Turn failure, warning, and success emails on or off independently for each pipeline. Keep the workspace defaults where they fit, or override them for pipelines that need different notifications.
Bring your first source into Supaflow.
Try Supaflow or walk through your use case with us.
See plans and pricing