n8n Google Drive to Google Sheets Receipt OCR with Gemini

Turn new Google Drive receipt images into structured Google Sheets rows—automatically

This n8n workflow watches a specific Google Drive folder for newly added receipt images, uses Google Gemini OCR/image analysis to extract key fields (receipt ID, table number, line items, total price), and then appends the data to Google Sheets—skipping duplicates by receipt ID.

What this workflow does

• Monitors every minute for new files in a configured Google Drive folder.
• Downloads the newly created receipt image from Google Drive.
• Extracts receipt data with Google Gemini by sending the image and requesting a JSON response containing:
  • receipt_id
  • table_number
  • line items (including quantities)
  • total_price
• Parses the returned JSON and formats the line items into a single comma-separated string.
• Prevents duplicate rows by checking Google Sheets for an existing row with the same receipt ID.
• Appends only new receipts to the target Google Sheets spreadsheet; if the receipt ID already exists, the workflow stops without changes.

Use cases

• Keep a live receipt log by dropping receipt photos into a shared Drive folder.
• For SaaS operators and operators teams: centralize receipt capture for reconciliation and reporting in Google Sheets.
• For automation engineers: build a reliable OCR-to-spreadsheet pipeline with duplicate detection.

Technical details

• Google Drive trigger: runs every minute when a new file is created in the watched folder.
• Google Gemini (Google PaLM): image analysis step to extract receipt fields as JSON.
• Google Sheets: searches for an existing receipt ID, then appends a new row if missing.
• Workflow logic using nodes such as if, code, no op, and supporting elements (e.g., a sticky note for guidance).

Setup requirements

Connect Google Drive OAuth2 credentials and select the folder to watch, add a Google Gemini API credential, connect Google Sheets OAuth2, and ensure your sheet columns match the workflow mapping (e.g., id, order, table, totalprice) so the duplicate check and append work correctly.