What is Optical Character Recognition (OCR)?
Optical Character Recognition: Technology that converts images of text, such as scans, photos, and PDFs, into machine-readable characters. It reads text but does not understand it.
Optical character recognition turns pixels into text. Given a scanned page or a photo of a receipt, it outputs the characters it finds, often with their positions on the page. Classic OCR engines like Tesseract and cloud services from major providers are fast and cheap, and they handle clean printed text well.
OCR alone does not know which number is the total or which date is the due date. Older systems added templates or rules per document layout to map text to fields, which broke whenever a vendor changed its invoice design. Today, multimodal LLMs can read the document image directly and return structured fields, with OCR often still used for searchable text and to cross-check numbers.
Why it matters when you build or buy AI
If a vendor pitches "OCR" as a document automation solution, ask what happens after the text is read. The hard parts are field mapping, validation, and handling poor scans, handwriting, and multi-page tables. For most finance workflows in 2026, you want IDP with model-based extraction and validation, not raw OCR with templates.
Example
OCR reads a crumpled taxi receipt as a block of text. An extraction step then identifies the merchant, the date, the total, and the tax amount from that text and the image.
Related terms
More in Finance Ops
Book 20 minutes, or send a brief
A former CTO replies. You leave with a written plan and a fixed estimate.
Prefer LinkedIn? Connect with Harish.