-
-
-
-
-
For whom?
-
-
-
-
With the help of OCR (Optical Character Recognition), Billit can automatically recognize and extract data from added PDF invoices. This speeds up the processing workflow and reduces manual data entry. In this section, you will learn how OCR works, what you can expect from it, and what you should take into account.
Although OCR technology has evolved significantly, it remains an automated process that is not infallible. Errors can occur for several reasons:
| Cause | Explanation |
|---|---|
| Poor document quality | Distorted scans, crumpled or skewed papers. Unreadable text due to low resolution. Handwritten notes |
| Unconventional or complex layouts | Some invoices are poorly structured or contain too many graphical elements. Information appears in unexpected locations (e.g. VAT number at the bottom of the page) |
| Language or formatting differences | Currency symbols, date formats, comma/decimal usage. Multilingual documents |
| Similar data mixed together | Multiple dates on the invoice (issue date, delivery date, due date). Multiple amounts (total, excluding VAT, advance payment, etc.) |
That is why it is always important to carefully check the proposed data and, where necessary, adjust it manually before processing the invoice.
