A document scanner beside a laptop running a Paperless-ngx archive

DOCUMENT WORKFLOW

Paperless-ngx won't import or OCR?

Follow one sample document from the consume folder through OCR and search, then verify the stored file and metadata.

  • One sample document through the full path
  • OCR text checked, not just a green status
  • Original file and metadata retained

THE THREE FAILURES WE SEE

What is going wrong?

01

The file never leaves consume

A PDF or image sits in the consume folder while the dashboard shows no new document.

02

OCR text is empty or wrong

The document appears, but search cannot find its text or the OCR layer is garbled.

03

Search cannot find the document

The correct correspondent or tag exists, but a known phrase returns no result.

HOW WE TEST

Find the failing layer, then prove the fix.

  1. Confirm the consume path. Check the container's consume mount and permissions. A file in the host folder is not necessarily visible inside the container.
  2. Read the task and worker logs. Find the document task and its failure message. Separate a file/permission error from an OCR language or dependency error.
  3. Test one clean document. Use a simple, text-based PDF or a clear scan, then wait for the task to finish before uploading a batch.
  4. Verify original, OCR, and search. Open the stored original, inspect the text layer, and search for a unique phrase from the page.
A scanned document and Paperless-ngx search result on a laptop
A successful import has a stored original, readable OCR text, and a search result.

SERVICE SCOPE

What is included

Included

  • One Paperless-ngx instance and one sample document
  • Consume-path and permission check
  • OCR/task log review
  • Original, text, and search verification

Not included

  • Scanning a whole physical archive
  • Repairing a damaged database
  • Custom OCR model training
  • Guaranteeing every handwriting style

The session covers one agreed device and one verified path. Anything outside the written scope is quoted separately before work starts.

FAQ

Before you book

Will you delete my documents?

No. The test uses one sample and leaves existing documents untouched.

Why does OCR fail on a clear scan?

The container may lack the language pack, the file may be image-only, or the task may have stopped before OCR. The task log distinguishes these.

Can you set up automatic tagging?

Basic matching rules can be reviewed, but this session focuses on a working import, OCR, and search path.

NEXT STEP

Tell us what is happening.

Send the device, software, exact symptom, and what you have already tried. We will confirm scope and a fixed quote before booking.

No passwords. A real reply comes from [email protected].

FROM THE BLOG

Paperless-ngx Setup Help Guides

All guides