Every receipt, bill, and statement that enters Hubdoc goes through our Optical Character Recognition (OCR) process. Computer software mines these documents for data, and then Hubdoc adds in some human quality assurance to certify that the correct information is extracted.
Extractor means no data entry or filing for you!
The date, amount, and vendor's name are extracted and stored in Hubdoc along with the document. If present, the invoice number and due date are also extracted.
The extraction process generally takes fewer than 24 hours. Documents waiting to undergo the extraction process can be found in the 'Processing' tab. Once Hubdoc has completed the extraction, documents will move to the 'Review' tab. If Hubdoc is unable to extract information, documents will move to the 'Failed' tab.
A document could fail the extraction process for a few reasons. The most common are:
- An image is too blurry, crumpled or faded for our OCR to properly recognize the information.
- A piece of mandatory information including the date, amount or vendor's name is missing from the image.
- There are multiple receipts in the same image. With the exception of multiple receipts for the same transaction (for example: the original invoice from the vendor and the credit card transaction with the tip), our OCR is unable to distinguish multiple receipts in the same image.