New input load types for BigQuery — smarter loading with Clone
We've improved how Storage tables are loaded into your BigQuery transformations and workspaces. BigQuery now supports the same smart load-type selection that Snowflake already uses — Keboola can pick the best method automatically for each input table.
What's new
Three loading methods are available, and an Auto mode picks the best one per table:
- View — the workspace points at the live Storage table, so you always read the current data. (Today's default for most projects.)
- Clone — the workspace gets its own independent, point-in-time snapshot of the table. Created instantly, uses virtually no extra storage, and behaves like a normal standalone table.
- Copy — a full physical copy of the data; used only when neither Clone nor View applies.
When Clone is the default, Keboola chooses automatically:
- Clone for a normal whole-table load from your own project.
- Falls back to View for tables that can't be cloned but can still be referenced live (linked/shared tables and aliases).
- Falls back to Copy when you filter rows or select/retype only some columns.
You can always override this per table with the Load type option in the input mapping (Auto / Copy / Clone / View).
Which method applies to which table
In practice the method is decided by the kind of table and what you ask Keboola to do with it:
- Clone — the standard case: a regular Storage table that lives in your own project, loaded as a whole table (no filters, no column selection or data-type changes, not incremental). The workspace gets an independent, point-in-time copy that behaves like a normal table.
- View — used for tables that can't be cloned but can still be read live: linked / shared tables (data shared into the project from another project), external tables, and alias tables. They're loaded as a read-only view and marked with a view icon in the input mapping.
- Copy — used whenever the load is not a plain whole-table load: you filter rows, select or retype only some columns, or load incrementally. Neither Clone nor View can do that, so Keboola makes a real physical copy.
In short: a whole-table load of your own data is cloned; a shared, external, or alias table becomes a live view; and any filtered, column-shaped, or incremental load is copied.
What's the default today?
Most BigQuery projects currently default to View — and that's perfectly fine. Nothing changes for you; your transformations and workspaces keep working exactly as before.
Want Clone as your default now?
Turn off the BigQuery Input Mapping – Default Load Type View feature in your project's Settings → Features. With it disabled, new input loads default to Clone (with the automatic View/Copy fallbacks above).
Heads-up — Clone becomes the default for everyone
We'll be gradually disabling the Default Load Type View feature across all BigQuery projects, completing by July 15, 2026. After that, Clone will be the default load type on BigQuery, bringing it in line with Snowflake.
Why Clone is better
- More predictable runs — each run works on a stable snapshot of its inputs instead of data that can change mid-run.
- Behaves like a real table, without view-specific limitations.
- Near-instant, with no meaningful extra storage cost.
If a workflow relies on always-live input data, you can keep using View per table at any time.