OCR: what is it?

Table of contents:

• What's the OCR?
• How you benefit from the OCR
• How it works
• Skip the OCR
• OCR has been canceled

What's the OCR?

When importing a PDF to your Billtobox account, it first goes through the optical character recognition (OCR) which extracts the data from the invoice's image. The extracted data will be used to create an electronic invoice.

Here's the data the OCR extracts from invoices:

• Customer or supplier
• Document type (invoice, credit note, expense)
• Invoice number
• Invoice date
• Due date
• Currency
• Total amount (VAT excl. and VAT incl.)
• VAT rate
• Reference
• IBAN

The OCR process is two-fold:

1. Computer recognition
2. Human recognition

This allows for greater accuracy of data because Billtobox does not rely solely on software, humans actually check and complement the software's work.

Note: to get better results, the OCR requires your documents to be scanned with a minimum resolution of 300 dpi.

How you benefit from the OCR

• Find any invoice in a matter of seconds.
  • Since the data has been extracted by the OCR, your invoices' data is organized and searchable. That wouldn't have been possible had your invoices remained in PDF format.
• If you're using accounting software to do the books, the electronic invoices created thanks to the OCR will be in UBL format that your software is able to read. No need to type each invoice's data manually.
• Pay your invoices without having to manually fill out your suppliers' IBAN and structured message.

How it works

Simply import an invoice to your account and it'll automatically get sent through the OCR. Once finished, your invoice gets a new coloured status, related to its due date.  

Skip the OCR

Don't want to wait for the OCR result?

You can choose to fill out the required data manually from the details page. 
This can come in handy when you're in a rush to forward a document to your accountant.

OCR has been canceled

If you accidentally uploaded a document that is not an invoice or if the document is unreadable, the OCR service may cancel the document.

In this situation, the document will get the "To accept" state but you'll notice that not all data has been filled out.

In the document's events, you'll find why the OCR was canceled.