Duplicating a Project

You can follow the steps below to create a copy of an existing project in the same or a different tenant.

To duplicate a project:

1. Go to Menu → Projects. The project cards are displayed.

2. Click the ellipsis icon on the project you want to duplicate and select Copy Project.

   • The ellipsis icon appears when you hover over the project card.

3. Review the component copy behaviour while copying the project with and without data in the dialog that appears. If you agree to the terms, proceed further.

4. Click Proceed to open the Copy Project window.

5. Specify a custom name for the copied project. If left blank, it will be named as a copy of the source project.

6. By default, the environment of the source project is selected and you cannot change it.

7. Select the workspace where you want to copy the project.

8. Select the copy options. Possible values:

   • With Only Metadata - Copies just the data pipeline structure without transferring any datasets. All blocks will remain in the unbuilt state, allowing users to build them afresh as needed—just like in the earlier version.

   • With data – Copies both the pipeline metadata and the associated datasets from the source project, ensuring a complete and ready-to-use duplicated project. Connectors and authentication credentials are also copied to maintain input and output connections.

9. Select DataApps and Prediction Services you want to copy from the source project to the duplicated project. Note that you cannot copy the scheduled DataApps.

10. By default, the secrets are copied.

    • If the project is being copied to a space workspace the platform reuses the existing secrets .

    • If the project is being copied to a different workspace the platform will automatically include the secrets by default to ensure compatibility.

11. Click Yes when copying a project, if there are name conflicts in components such as projects, artifacts, models, environments, connectors, or DataApps, the platform will notify the user before proceeding. This allows the user to review and take appropriate action.

    • If the user chooses to proceed despite the conflicts, recipe templates within the copied project may require updates to their input/output references to ensure smooth end-to-end execution.

    • For rapid model recipes, output names are always auto-renamed during the copy process—regardless of whether the destination is the same or a different workspace—to maintain uniqueness.

    • In cases where name conflicts are accepted, the new names will retain the original name with a sequence number appended for easy identification.

Once the copy action is complete, the duplicated project will appear on the Projects page of the selected workspace. You can track the progress and view updates through the Notifications panel.

Notes

• When duplicating a project, all components—including Prediction Services, Environments, and DataApps (excluding Job-type DataApps)—are seamlessly copied. You can also selectively include DataApps and Prediction Services by checking the corresponding checkboxes during duplication.

• Chat conversations within AI-Assisted Recipes are also duplicated, preserving context and interactions. To view duplicated chats, navigate to the AI-Assisted Recipe page and click Refresh Chat to rerun queries and retrieve updated results.

• Upon importing components into a new project, they will initially appear in an unbuilt state:

  • For datasets imported via local files, use Add File to upload the dataset (accessible by right-clicking on the dataset block).

  • For datasets imported through connectors, use the Reload option (accessible by right-clicking on the dataset block).

• When you run the duplicated project, all components in the data pipeline will be assigned a new name.

• Fivetran connectors are not supported in the Copy Project feature.