What this workflow does
This workflow takes exported Google Keep notes stored in a Google Drive folder and finds only the ones about expenses.
It reads each note, gets the spending amounts in euros using an AI model (OpenAI GPT-4o-mini), and cleans the data with readable dates.
Finally, it adds or updates that data into a Google Sheet for easy budgeting and review.
Who should use this workflow
This is for people who use Google Keep to track expenses but spend too much time copying data manually.
It helps automate saving time and avoiding mistakes when moving notes into budgets.
Tools and services used
- Google Drive: Stores the exported Google Keep JSON note files.
- Google Sheets: Saves the cleaned and structured expense data.
- OpenAI API: Uses GPT-4o-mini model to pull euro amounts from note text.
- n8n: Runs the automated steps connecting all services.
How the workflow processes data
Inputs
The workflow starts by reading all Google Keep JSON files in a specified Drive folder.
Processing Steps
- Split files into small batches of 10 to manage processing load.
- Filter only the files ending with a .json extension.
- Download each JSON file and parse the note content.
- Keep only notes containing keywords for spending in French like “dépensé” or “depense”.
- Exclude archived notes by checking the archived flag.
- Send the note text to OpenAI GPT-4o-mini through a Langchain node to extract exact euro amounts.
- Convert timestamps to human-readable creation and edit dates.
- Prepare the data fields ready for export including the AI extracted amount.
Output
Data is appended or updated in a chosen Google Sheet, thus creating a clear spreadsheet track of expenses.
Beginner step-by-step: How to run this workflow in n8n
Import and setup
- Download the full workflow file using the Download button on this page.
- In your n8n editor, use the “Import from File” option to load the workflow.
- After import, open the workflow to configure credentials for Google Drive, Google Sheets, and OpenAI API.
- Update any folder IDs, spreadsheet IDs, or sheet names in the respective nodes that fit your Google Drive and Sheets setup.
Testing
- Run the workflow manually by clicking the Manual Trigger node called “When clicking ‘Test workflow’”.
- Check that the expense data appears in your Google Sheet as expected.
Activate for production
- Once testing works, activate the workflow in n8n for automatic or scheduled runs.
- If needed, add a Cron node before the Drive search step to schedule routine processing.
- Monitor successful runs and error logs inside n8n.
Use self-host n8n if you prefer to run n8n on your own server for better control.
Input and Output Details
Input: Collection of Google Keep JSON notes inside one Google Drive folder.
Output: Structured spreadsheet rows with the original note, creation and edit dates, archive status, and extracted euro amounts.
Customization ideas
- Change the expense keyword filter in the “Filter” node to include other terms like “invoice” or “payment”.
- Select a different AI model in the “OpenAI Chat Model” node to balance cost versus accuracy.
- Adjust the batch size in the “Loop every 10 items” node to control API usage.
- Add more note details like labels or location in the export by expanding the fields in the “Set the fields for export” node.
- Skip AI extraction completely by removing or disabling the AI node for raw text export only.
Possible issues and fixes
- Issue: No files found in Drive folder.
Fix: Check the folder ID in the “Search in ‘Keep’ folder” node is correct. - Issue: AI returns empty or wrong amounts.
Fix: Adjust the AI prompt or try a better GPT model like GPT-4. - Issue: Google Sheets node fails to add data.
Fix: Confirm Google Sheets ID and permissions are correct and the connected account has access.
Summary
✓ Automatically reads and filters Google Keep exported notes about expenses.
✓ Extracts euro amounts from note text using OpenAI AI.
✓ Converts timestamps and adds data into Google Sheets.
→ Saves hours on manual data entry.
→ Reduces errors in expense tracking.