How it works
When a document is imported, Formality checks whether it already has text. Scanned pages are sent to OCR ("Scanning" in Imports) and the text is stored page by page with the document. In the viewer the recognised text sits as an invisible layer over the original page, so the scan keeps its stamps and signatures while you search it or select and copy a passage.
From then on a scan behaves like any other file. The text is indexed for search, the AI extracts fields from it, and each value or answer links back to the page it comes from.
Signature pages are often the hardest part: a handwritten date next to a signature. When you reprocess an agreement you can turn on a targeted re-OCR of pages where the signature date is handwritten or unreadable.
What you get
- OCR runs automatically on import
- Search inside the document, with highlights
- Extraction and citations work on scans
- Targeted re-OCR of handwritten signature dates
Why teams use it
Concessions, site leases or supply contracts signed on paper years ago become usable data.
Text recognition is part of the import. Nothing to install or run beforehand.
Find a clause in a 1998 scan the same way as in a recent Word export.
Extracted values on scans still point to the page they come from.
A dedicated pass reads handwritten dates that drive terms and deadlines.
More in Repository & import
Central contract repository
Every agreement, with its amendments and annexes, in one list.
Learn more →Bulk upload
Upload hundreds of files or whole folders at once.
Learn more →Inbox with document triage
Imported documents are triaged: accepted, uncertain or out of scope.
Learn more →Amendments and annexes linked to the contract
Each amendment and annex sits with its main agreement.
Learn more →