Extract and Organize Receipt Data for Expense Tracking with VLM Run and Google

This n8n workflow automates the process of extracting data from receipt images uploaded to Google Drive and organizing it into a Google Sheet for expense tracking. It leverages a Vision Language Model (VLM) for data extraction and Google services for storage and organization.

What it does

This workflow simplifies expense tracking by:

1. Monitoring Google Drive: It continuously monitors a specified Google Drive folder for new files (receipt images).
2. Extracting Receipt Data: When a new receipt image is detected, it uses a VLM (Vision Language Model) to extract key information such as vendor, date, total amount, and currency.
3. Preparing Data: It processes the extracted data, ensuring it's in the correct format for the Google Sheet.
4. Storing in Google Sheets: The extracted and formatted receipt data is then appended as a new row to a designated Google Sheet.
5. Archiving Receipts: After processing, the receipt image is moved to an "Archived" folder in Google Drive to keep the input folder clean and track processed items.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Account: Self-hosted or cloud.
• Google Account: With access to Google Drive and Google Sheets.
  • Google Drive Credential: Configured in n8n for both the "Google Drive Trigger" and "Google Drive" nodes.
  • Google Sheets Credential: Configured in n8n for the "Google Sheets" node.
• VLM Run Node: This workflow is designed to work with a VLM (Vision Language Model) node for data extraction. While the specific VLM node is not present in the provided JSON, it is a crucial component implied by the workflow's purpose. You will need to add and configure a VLM node (e.g., an OpenAI Vision node, or a custom VLM integration) capable of processing images and extracting structured data.

Setup/Usage

1. Import the workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Google Drive Trigger:
   • Select your Google Drive credential.
   • Specify the Google Drive folder where you will upload your receipt images.
3. Configure the VLM Node (Not in JSON, but required):
   • Add a suitable VLM node (e.g., OpenAI Vision, or another image-to-text/data extraction service).
   • Configure it to take the image data from the Google Drive Trigger as input.
   • Ensure it's prompted to extract vendor, date, total_amount, and currency from the receipt image.
4. Configure Edit Fields (Set) Node: This node is used to standardize the data extracted by the VLM. Adjust the expressions to map the VLM's output fields to the desired column names for your Google Sheet (e.g., {{ $json.vlm_output.vendor }}, {{ $json.vlm_output.date }}, etc.).
5. Configure Google Sheets Node:
   • Select your Google Sheets credential.
   • Specify the Spreadsheet ID and Sheet Name where your expense data will be stored.
   • Ensure the "Operation" is set to "Append Row".
   • Map the fields from the "Edit Fields (Set)" node to the correct columns in your Google Sheet.
6. Configure Google Drive (Move File) Node:
   • Select your Google Drive credential.
   • Set the "Operation" to "Move File".
   • Specify the "File ID" to be the ID of the file from the Google Drive Trigger.
   • Specify the "Destination Folder ID" for your "Archived Receipts" folder in Google Drive.
7. Activate the workflow: Once all nodes are configured, activate the workflow.

Now, any receipt image you upload to the specified Google Drive input folder will be automatically processed, its data extracted, added to your Google Sheet, and then archived.