Product
Invoice OCR that returns structured, editable invoice data
Drop in a PDF, scan or phone photo of an invoice. The document is read in your browser, every field comes back with a confidence value, and you can correct anything before exporting to Excel, CSV or JSON.
Try it now
Upload your own file. Text is recognised in your browser and the document is not uploaded. Complex layouts may need manual review before you export.
Check the result before you export
Drop a file here, or click to choose one
PDF, JPG or PNG · 20 MB
What Invoice OCR does
Invoice OCR is optical character recognition combined with document understanding. Plain OCR gives you a wall of text; invoice OCR also decides which piece of text is the invoice number, which is the due date, and which numbers belong to which line item.
That distinction is what makes the output usable. Instead of retyping an invoice into a spreadsheet, you review a structured record where each field is already filled in and every value stays editable.
Low-confidence values are highlighted, so review time goes to the fields that actually need a human rather than to the whole document.
What gets extracted
Vendor and buyer
- Vendor name
- Vendor address
- Vendor tax / VAT number
- Buyer name and address
- Contact details when printed
Document
- Invoice number
- Purchase order reference
- Invoice date
- Due date
- Payment terms
- Currency
Amounts
- Subtotal
- Discount
- Tax rate
- Tax amount
- Shipping
- Total due
- Amount paid / balance
Line items
- Description
- Quantity
- Unit
- Unit price
- Line discount
- Line tax
- Line total
How it works in practice
Drag, drop or browse
Drop a file onto the upload area or pick it from your device. Type and size are checked before anything is processed.
Progress you can see
Preparation, reading and field extraction are reported as distinct stages, so a slow document never looks like a frozen page.
Side-by-side review
The original document sits next to the extracted fields, so you can verify a value without opening the source file separately.
Editable line items
Add, remove, reorder and correct rows. Totals recalculate as you edit, so the exported figures stay internally consistent.
Three export formats
Excel (.xlsx) with a summary sheet and a line-item sheet, CSV for imports, and JSON for systems that consume the raw record.
Honest failure states
Unsupported file, oversized file, unreadable scan and empty result each produce a specific message and a way forward.
Who it is for
Accounting teams
Bookkeepers processing supplier invoices for several clients, who need a consistent export rather than a scanned PDF archive.
Finance and AP
Accounts payable staff matching invoices against purchase orders and preparing payment runs.
Small business owners
Anyone who currently types invoice totals into a spreadsheet by hand at the end of each month.
Supported formats
- PDF — native (text-based) and scanned, single or multi-page
- JPG / JPEG — photos and scans
- PNG — screenshots and exports
- Up to 20 MB per file
How your documents are handled
- Recognition runs inside your browser. The document is not uploaded to a server to be read.
- Because the file stays on your device during extraction, its contents are never written to our logs or held in a processing queue.
- A conversion record — filename, size and page count — is stored against your account so your allowance and history are accurate. The document's contents are not part of that record.
- The connection to the application is HTTPS/TLS only.
- You can delete a conversion record at any time from your account.
Full detail, including what we deliberately do not claim, is on the Security page.
Frequently asked questions
Ready to try Invoice OCR?
Upload a file above, or read the documentation to see how the whole workflow fits together.