Importing Data
Vault is designed to handle your financial data with maximum privacy. The importing process is the foundation of how Vault analyzes your expenses without ever compromising your security.
How Importing Works
Unlike traditional financial apps, Vault does not connect to your bank accounts via third-party APIs (like Plaid). Instead, it relies on local file imports to ensure your data stays under your control.
1. File Selection
You can import your data by dragging and dropping files directly into the Vault interface or by using the file picker. Vault supports Bank Statements (PDF), Transaction Exports (CSV), and Spreadsheets (XLSX/TSV).
2. Local Processing (On-Device)
Once a file is selected, the processing happens entirely on your Mac. Vault's local parser reads the file content, and categorization happens using local AI models. No data is ever uploaded to a cloud.
3. Data Storage
After the import is complete, the structured data is stored in a local database on your machine at ~/Library/Application Support/Vault/.
Best Practices
- Direct Exports: Always prefer CSV exports over PDFs when available, as they provide higher accuracy.
- Clean Files: Ensure your CSV files have headers (e.g., "Date", "Description", "Amount").
- Sensitive Data: You don't need to scrub your files before importing; they never leave your computer.
Troubleshooting
Unsupported Format: If a specific bank's PDF format isn't being recognized, try exporting a CSV instead.
Incorrect Categorization: You can manually override any category after the import is finished. Vault learns from these corrections for future imports.
Last modified: January 10, 2026