Invoice OCR and data extraction
Upload vendor invoices as PDFs, scans or phone photos. Get the vendor, invoice number, dates, totals and every line item as a CSV, or send the invoice to QuickBooks Online or Xero as a bill.
Or drop them herePDF, JPG, PNG or HEIC, up to 20 at once
- Vendor
- Ink + Paper Co.
- Invoice number
- INV-1047
- Invoice date
- 2026-03-08
- Due date
- 2026-04-07
- Subtotal
- 356.15
- Tax
- 28.85
- Total
- 385.00
- Currency
- USD
Invoice line item extraction
Every line comes back as its own row, with description, quantity, unit price and amount in separate columns. A line that runs across a page break is joined back into one row.
| Description | Quantity | Unit price | Amount |
|---|---|---|---|
| Letterpress business cards, 500 | 2 | 74.50 | 149.00 |
| Kraft A7 envelopes, box of 100 | 3 | 18.25 | 54.75 |
| Recycled copy paper, case | 4 | 32.60 | 130.40 |
| Delivery | 1 | 22.00 | 22.00 |
vendor_name,invoice_number,invoice_date,description,quantity,unit_price,amount Ink + Paper Co.,INV-1047,2026-03-08,"Letterpress business cards, 500",2,74.50,149.00 Ink + Paper Co.,INV-1047,2026-03-08,"Kraft A7 envelopes, box of 100",3,18.25,54.75 Ink + Paper Co.,INV-1047,2026-03-08,"Recycled copy paper, case",4,32.60,130.40 Ink + Paper Co.,INV-1047,2026-03-08,Delivery,1,22.00,22.00
{
"vendor_name": "Ink + Paper Co.",
"invoice_number": "INV-1047",
"invoice_date": "2026-03-08",
"due_date": "2026-04-07",
"subtotal": "356.15",
"tax_total": "28.85",
"total": "385.00",
"currency": "USD",
"line_items": [
{ "description": "Letterpress business cards, 500", "quantity": "2", "unit_price": "74.50", "amount": "149.00" },
...Matches the printed total
How to extract data from an invoice
Upload
Drop PDFs, scans or phone photos. A 12-page invoice is read as one invoice, not 12.
Check
Read the header fields and line items. If subtotal plus tax does not equal the total, the invoice is flagged so you can fix it.
Export or send
Download the line items as CSV or JSON, or send the invoice to QuickBooks Online or Xero as a bill.
Send invoices to QuickBooks Online and Xero as bills
Connect QuickBooks Online or Xero and the checked invoice becomes a bill, so nobody types it in again.
Sending to QuickBooks Online and Xero is on Pro plans. QuickBooks details
- Vendor
- Ink + Paper Co.
- Bill number
- INV-1047
- Bill date
- 2026-03-08
- Due date
- 2026-04-07
- Expense account
- Office supplies
- Amount
- 385.00
Invoice OCR for scans, photos and long invoices
Off by 45.00. Check the total on page 2 before you export or send this invoice.
Invoice to CSV, JSON, QuickBooks or Xero
Line item CSV
One row per line item, with the vendor, invoice number and date on every row. Opens in Excel and Google Sheets.
CSV export detailsJSON
The header fields, line items and check result, for your own code or the API.
JSON export detailsQuickBooks Online
A bill with the vendor, dates, amount and the invoice PDF attached.
QuickBooks detailsXero
A bill under the right contact, with the source file attached.
Who uses invoice data extraction
Bookkeepers and accountants
Clear a client's stack of vendor invoices into QuickBooks Online or Xero. Clients can send invoices to you through your portal.
LedgerBox for firmsBusiness owners
Stop typing bills by hand. Upload the month's invoices, check the totals and send them to your books.
Accounts payable in QuickBooksDevelopers
Send invoices through the API and get JSON back, with the totals check already run.
API and SDKWhat LedgerBox does not do with invoices
It gets the data out and into your books. It does not route invoices for approval or pay them. For that, use your accounting software or an accounts payable tool, and let LedgerBox handle the typing.
Is it safe to upload invoices?
Invoice extraction questions
What is invoice OCR?
Invoice OCR reads the text on an invoice image or PDF. LedgerBox goes a step further and turns that text into fields: vendor, invoice number, dates, totals and each line item, with the totals checked.
How do I extract data from a PDF invoice?
Upload the PDF, check the header fields and line items LedgerBox finds, and download the CSV or JSON. Your first 10 pages are free.
Which fields does it extract?
Vendor name, invoice number, invoice date, due date, subtotal, tax, total and currency. Each line item comes back with description, quantity, unit price and amount.
Can I convert an invoice to Excel?
Yes. Download the line item CSV and open it in Excel or Google Sheets. Each line item is one row.
Can I import invoices into QuickBooks Online?
Yes. On Pro plans, connect QuickBooks Online and send each checked invoice as a bill. The vendor is found or added, and the invoice PDF is attached.
Does it work with Xero?
Yes. Connect Xero and send invoices as bills under the right contact, with the source file attached.
Does it work with scanned or photographed invoices?
Yes. Scanned PDFs and JPG, PNG and HEIC photos are read with OCR. Lines that are hard to read are flagged for you to check.
What happens if the totals do not add up?
If subtotal plus tax does not equal the total within a cent, the invoice is flagged. You fix the number or confirm it before you export or send it.
Can I upload many invoices at once?
Yes. Upload up to 20 files at a time, up to 8 MB each. A multi-page invoice is read as one invoice.
Will the same invoice be sent to my books twice?
No. Before it sends a bill, LedgerBox checks for one already in QuickBooks Online or Xero and stops if it finds a match.