Connect Google Drive to Rig

How the connection works

Rig provisions a service account for your workspace: a Google identity owned by Rig with no access to anything until you give it some. You share a folder with its email address, exactly as you would share with a colleague, and Rig can then read that folder and nothing else.

That makes the blast radius easy to reason about. Rig sees the folders you shared, the permission is visible in your own Drive, and unsharing revokes access immediately without involving us.

Before you start

• Permission to share the folder. If it belongs to someone else, they need to do the sharing.
• A folder rather than scattered files. Rig can take individual files, but a folder means new files are picked up automatically as they are added.
• Some Workspace domains block sharing outside the organisation. If yours does, an admin will need to allow it for this address before the share will stick.

Generate your Rig service account

The service account is generated once for your whole workspace, not once per connector. If you have already generated one for another Google source, reuse that same email address and skip to the next section.

1. In Rig, open Connections and go to Sources & actions.
2. In the Google service account card at the top of the page, click Generate service account.
3. Copy the email address Rig gives you. It ends in .iam.gserviceaccount.com.

Connections → Sources & actions: the Google service account card sits above the connector grid.

Share the folder with Rig

1. In Google Drive, right-click the folder and choose Share.
2. Paste the service-account email and set it to Viewer. Rig only ever reads.
3. Turn off the notification email if Google offers it. Nobody reads a service account's inbox.
4. Share.

Connect it in Rig

1. In Rig, click Check for new shares. Rig asks Google what the service account can now see and lists what it finds, marking anything already connected so you do not add it twice.
2. Pick the folder and connect it.
3. To pin the source to specific folders instead, paste one or more folder IDs into the source's Folder ID(s) field, comma-separated. The ID is the last part of the folder URL, drive.google.com/drive/folders/<ID>. Leaving it blank scans everything shared with the service account.
4. Click Test, then Sync.

What Rig pulls in

Drive lands in its own google_drive schema. Two kinds of file are handled, and they land differently:

• Tabular files: CSV, Parquet, JSONL and Excel become queryable tables
• PDFs: consolidated into a single documents table, one row per page, so the text is searchable

How files land in the warehouse

Tabular files are grouped by shape, not by file. Twelve monthly exports with the same columns become one table with twelve months of rows, rather than twelve near-identical tables. Every row carries the file it came from, so you can still filter back to a single export.

This matters more than it sounds. One customer's invoice folder held 285 files across about 33 recurring shapes. Landing one table per file produced 888 tables, 407 of them with fewer than 20 rows, which is unusable for both people and AI tools. Grouping by shape turns the same folder into a few dozen tables that mean something.

PDFs are versioned rather than overwritten. When a file changes, Rig appends a new version instead of replacing the old one, so history is preserved.

Troubleshooting

• Check for new shares finds nothing: the share did not land. Open the folder's Share dialog and confirm the service-account address is listed. Workspace domains that restrict external sharing will silently drop it.
• Your spreadsheets are missing: they are probably Google Sheets rather than uploaded files. Use the Google Sheets connector for those.
• Fewer tables than files: that is the by-shape grouping working. Check the source file column before assuming something is missing.