Documents: OCR and power features

Search the text inside images with OCR, recover deleted files from the 30-day trash, plus audit logs, storage quotas, duplicate detection, and watermarks.

OCR reads the text inside pictures and turns it into searchable text — photograph a signed delivery slip, and later searching the customer name finds that photo. This page also covers the trash, audit logs, storage quotas, duplicate detection, and watermarks.

OCR full-text search#

After an image is uploaded, the system periodically sends new images to the OCR service; the recognized text goes into the search index. Afterwards, typing a keyword from inside the picture (a customer name, a document number) in the search box finds that image.

1. The image is uploaded to Documents.
2. A scheduled job dispatches new images to the OCR service.
3. Recognition results are written back, and the image becomes searchable.

30-day trash#

• Deleted files go to the trash first and stay for 30 days — restore any of them with one click during that window.
• After 30 days they are permanently deleted and cannot be recovered.
• In a hurry to free space, you can also delete permanently by hand — think twice before you do.

Audit log and storage quota#

• Full audit log of every action: who uploaded, downloaded, shared, or deleted which file and when — everything is traceable.
• Tenant storage quota: each tenant has a storage cap; check usage on the stats page and ask your admin to expand it before it fills up.

Duplicate detection and watermarks#

• Duplicate detection: if an uploaded file is identical in content to one that already exists, the system warns you so you do not waste space on copies.
• Watermarks: a watermark can be stamped automatically on download or share to discourage casual forwarding. (What the watermark shows follows the actual page.)