Short answer: OCR (optical character recognition) reads the text inside a photo or scan of a document and turns it into text a computer can search and copy. Extracting data goes one step further — pulling key details like an invoice number, date or amount into fields, so your documents become searchable records instead of flat pictures.
OCR, in one minute
When you photograph a bill, your phone captures a picture — to a computer it's just coloured dots, with no idea what the words say. OCR is the step that recognises those dots as letters and numbers, turning the image into real, selectable text. That's the difference between a photo of an invoice and an invoice you can actually search.
From "text on the page" to "data you can use"
Recognising the text is half the value. The other half is structure: identifying which words are the invoice number, which is the date, which is the party name or amount, and putting them into fields. Once a document's key details sit in fields, you can filter and sort records — "show me everything from this supplier", "find the bill for ₹48,000" — without ever retyping. ScanPix does this capture-and-fill automatically; see automatic data extraction.
Why it makes everything searchable
A folder of scans is only as findable as its filenames. With the text recognised, you can search by what's written inside a document — a name, a number, a single word — and land on the right page in seconds. That's what turns a pile of PDFs into a real, searchable archive. More on how records are organised in document management.
Where businesses use it every day
Invoices and bills (find any one by number or amount), agreements (search by clause or party), ID and KYC documents (collect and locate by name), delivery notes and forms. Anywhere you currently flip through paper or scroll a camera roll to find one document, extracting the data removes the flipping. Professional and financial teams lean on this hardest — see professional services.
A note on accuracy and privacy
Recognition is very good on clean captures and weaker on blurry or skewed photos — good edge detection and lighting help. Just as important: the documents being read are yours. They're encrypted in transit and at rest and stay private to you and your team. Read the security page for how your data is protected and how you keep ownership of it.
Frequently asked questions
What is OCR?
OCR (optical character recognition) is technology that reads the text inside a photo or scan of a document and turns it into text a computer can search and copy — so a picture of an invoice becomes data you can find and reuse.
How does extracting data from documents work?
After a document is captured, the text is read automatically and key details — like an invoice number, date, party name or amount — can be pulled into fields. Those fields make the document searchable and let you filter records without retyping anything.
Why does OCR make documents searchable?
A plain photo is invisible to search. Once the text is recognised, you can search by what is written inside a document — a name, a number, a word — instead of relying on the filename.
Turn your documents into searchable records
Capture a document and watch the details fill in — free on iOS, Android and the web.