Google Drive to Snowflake: How to Move Your Data
Move Google Drive files into Snowflake with Airbyte. Why Google Docs and Sheets are converted on export, folder sharing, and writing glob patterns per format.
Moving Google Drive into Snowflake turns a shared folder into queryable tables. Partner extracts, exported reports and files somebody drops in every week are invisible to your warehouse until something loads them, and Drive is where a great deal of that material already sits.
One distinction decides whether this pipeline does what you expect. A CSV or Parquet file in Drive is a file and arrives as rows. A Google Doc or Sheet is not a file in the same sense, and what reaches Snowflake is a conversion rather than the structured data you may be picturing.
Google Drive to Snowflake at a glance:
Why move data from Google Drive to Snowflake?
Two situations account for most of these pipelines.
The first is a file feed that happens to arrive in Drive rather than on an SFTP server. A partner shares a folder, a system exports to it nightly, and somebody currently downloads those files and uploads them somewhere else by hand. That is a pipeline waiting to be automated.
The second is joining that material to warehouse data, such as matching a vendor's product list against your own catalogue. If the folder holds Google Docs rather than data files, read the section on native formats before planning anything, because the destination may be wrong.
What do you need before you start?
One of these is a task for somebody else, and it is the one that decides whether anything appears at all:
Authentication, and a folder shared with it. OAuth is recommended on Airbyte Cloud and simplifies the setup considerably. On self-managed deployments you create a service account key, and then somebody must give that service account access to the folder, exactly as they would a colleague. The Google Drive source documentation covers both routes.
The folder link, copied from the Drive interface. Navigate to the folder you want and copy the current URL. The connector pulls from that single folder and includes every subfolder recursively, so choosing a root too high up sweeps in far more than you intended.
A glob pattern per stream. Format is configured per stream rather than per file, so a folder containing several file types needs patterns that select only the files matching each format. Patterns are relative to the root folder you specified.
Snowflake objects and a role. A warehouse, database, schema and a role able to create tables in that schema. Airbyte creates one table per stream with its own metadata columns alongside your data.
If your organisation restricts inbound traffic by IP, add the Airbyte Cloud IP addresses to the allow list on the Snowflake side before you begin.
How do you build a Google Drive to Snowflake pipeline in Airbyte?
Step 1: Look at what is actually in the folder
Open the folder and note the file types, because that determines everything downstream. A folder of CSV exports is a straightforward structured pipeline. A folder of Google Docs and Sheets is a document extraction problem, and the two have little in common beyond both living in Drive.
Step 2: Configure the Google Drive source
Click Sources in the left navigation, then New Source, and select Google Drive, following adding a source. Authenticate, paste the folder link, then define a stream for each file format you need with a name and a glob pattern selecting only files of that type.
Step 3: Configure the Snowflake destination
Click Destinations, then New Destination, and select Snowflake, following adding a destination. Supply the account identifier, warehouse, database, schema and role. The smallest warehouse is usually ample, since file feeds are modest by warehouse standards.
Step 4: Create the connection and schedule against arrivals
Click Connections, then New connection, select your streams and an incremental sync mode so new files are picked up without re-reading what has already been processed. Set the frequency against how often files actually land rather than how fresh anyone would like the data.
Check the resulting table after the first sync, particularly with CSV, since that format carries no type information and the connector has to infer what each column contains.
What happens to Google Docs and Sheets?
They get converted on the way out, because Google-native files are not stored as files in the ordinary sense. Google Docs are exported as Docx to preserve as much formatting as possible, and other exportable Google formats go through PDF. What reaches your pipeline is that export rather than the original.
The consequence catches people out with Sheets specifically. A Google Sheet looks like a table, so it is natural to assume it arrives as rows in Snowflake. It does not, because it is handled as a document rather than as structured data, and the result is text rather than typed columns.
If your folder is full of Sheets, the Google Sheets connector is the right tool and reads them as tabular data directly. Use this connector for genuine files, and use document file type deliberately when you want the prose, which is more often a vector store use case than a warehouse one.
Why does the folder choice matter so much?
Because one source covers one folder, and that folder includes every subfolder beneath it. Point the connector at a team's whole Drive and you have swept in everything anybody has ever filed there, including material nobody intended to put in a warehouse.
The opposite failure is quieter. On a service account the folder has to be shared with it explicitly, so a folder nobody shared produces an empty sync rather than an error, and a new subfolder created outside the shared root simply never appears. Neither shows up as a problem in your monitoring.
So agree a dedicated folder for anything feeding a pipeline, tell whoever drops files there that the location is now load-bearing, and use glob patterns to narrow rather than relying on the root to do it. That conversation costs ten minutes and prevents both failures.
Frequently asked questions
Will a Google Sheet arrive as a table in Snowflake?
No. Google-native files are converted on export, with Docs becoming Docx and other formats going through PDF, so a Sheet is handled as a document. Use the Google Sheets connector for tabular data.
Why is my sync returning nothing?
On a service account, check the folder has actually been shared with it. Then check your glob pattern, since it is relative to the root folder you specified.
Can one source cover several folders?
No. It pulls from a single folder, though every subfolder beneath it is included recursively. Create a source per folder, or nest them under a shared root.
Is there a file size limit?
The document file type format has a maximum of 1GB per file. Structured formats such as CSV and Parquet are governed by your sync resources rather than that ceiling.
Can I do this without writing code?
Yes. The setup above is entirely UI-driven, including the glob patterns. You may want Snowflake views casting CSV columns, since that format carries no type information.
Get your Google Drive data into Snowflake
Look in the folder first, because a folder of CSVs and a folder of Google Docs are different projects. Agree a dedicated folder and share it explicitly, write glob patterns that select only the files matching each format, and use the Google Sheets connector if what you actually want is tabular data.
Airbyte's connector catalog includes 700+ pre-built connectors, so file feeds can reach a warehouse alongside the systems they describe. For a similar file source, see SFTP to PostgreSQL, and for another Google source in a warehouse, Google Workspace Admin to BigQuery.
Integrate with 700+ apps using Airbyte
Move data from 700+ sources into warehouses, lakes, and beyond. Set up pipelines in minutes with pre-built connectors and the Connector Builder.