Use cases
Where Document AI saves the most time
Four situations the platform was designed around, and what it does in each.
Accounts payable
Finance teams receiving supplier invoices by email and post.
The problem
Keying invoice numbers, dates, line items and tax into the books by hand is slow and error-prone.
With Document AI
- Upload invoices in bulk or send them through the API from your inbox tooling.
- Line items, totals and tax are extracted and checked arithmetically.
- Publish approved invoices as bills, with the supplier matched to your existing contacts.
Expense management
Companies reimbursing employee receipts.
The problem
Receipts arrive as phone photos in every layout and language.
With Document AI
- Receipts are classified and their merchant, date, total and tax read from the image.
- Uncertain fields are routed to a reviewer instead of guessed.
- Approved receipts become expenses in your accounting system.
Bookkeeping practices
Accountants keeping the books of many clients.
The problem
Each client's documents must stay separate and land in that client's ledger.
With Document AI
- One client workspace per client, isolated from the others down to the database.
- Each workspace has its own connections, members and settings.
- Usage and cost are reported per workspace.
Bank statement processing
Teams reconciling accounts or assessing credit applications.
The problem
Statements are long, multi-page tables that copy badly out of PDFs.
With Document AI
- Transaction tables are reconstructed row by row across pages.
- Debits, credits and balances are kept as printed and exported to CSV.
- Results are available as JSON through the API for your own systems.
Process your first documents today.
Create an organisation, try the bundled sample documents, and call the API with your own key — in a few minutes.