Skip to main content
Duplicating clones a dataset — its rows, columns, and metadata — under a new name. It’s useful for:
  • Creating a variant before applying different transformations downstream
  • Forking a dataset before a risky edit, so you can compare side by side
  • Spinning up test data from a verified dataset
1

Open Manage Custom Data

From the Data Query widget’s Custom Data source, click Manage next to the Dataset label.
2

Choose Duplicate

Select the dataset and choose Duplicate.
3

Name the copy

Enter a name for the new dataset. The copy is created with that name.
Widgets continue pointing at the original dataset — duplicating doesn’t re-wire anything. To point a widget at the copy, open the widget and select the new dataset.

​
API

POST /v1/custom-data/{id}/duplicate clones the source dataset. Send new_name for the copy. The response returns the new dataset id.