Optical Character Recognition (OCR)

tl;drOptical character recognition is the technology that converts an image of text, a scanned invoice, a photographed receipt, a PDF, into machine-readable data.

It's the first step in most invoice-automation pipelines: before a system can code or match an invoice, it has to be able to read it.

OCR alone doesn't understand what it reads. It's often paired with AI that interprets the extracted text: recognizing that a string of digits is an invoice number rather than a phone number, or that a line item maps to a specific GL account. On its own, OCR is a decades-old technology; what's changed is how much intelligence now sits on top of it.

Back to the glossary