Paperless-ngx
Key Info:
Paperless-ngx is an open-source, self-hosted document management application that turns scanned and digital files into a searchable archive. Its web interface combines OCR, correspondents, document types, tags, storage paths, custom fields, permissions, workflows and versioned document files.
Details
Ingestion and OCR
Paperless-ngx can consume documents from an upload, watched folder, email rules and API-oriented workflows. It performs OCR when needed, can create an archival PDF/A copy and preserves the original file. Automatic matching can suggest correspondents, document types, tags and storage paths based on content. A deployment test should include native PDFs, photographs, poor scans, password-protected files, duplicates and a failed OCR task so operators understand the exception queue.
Organisation and search
Documents can be described with correspondents, document types, tags, custom fields, dates and storage paths. Full-text and advanced search use extracted content and metadata, while saved views can place dynamic collections on the dashboard. The document detail view supports preview, editing, history, download and share links. Versioned files preserve earlier underlying content while the main document keeps its metadata. Buyers should define a limited taxonomy before importing a large archive.
Permissions and workflows
Global and object-level permissions control which users and groups may view or edit documents and related objects. Workflows can react when consumption starts, a document is added or updated, or a schedule is reached, then assign or remove metadata, permissions and ownership or invoke email and webhook actions. These controls are powerful but require testing of private objects, workflow order, service credentials and the effect of changes on existing documents.
Fit and evaluation
Paperless-ngx may suit individuals, small organisations and technically capable teams that want a searchable self-hosted archive and accept responsibility for infrastructure and support. It should not be assumed to provide every enterprise records, legal-hold or vendor-SLA function. Evaluate installation, updates, identity, permissions, OCR languages, storage, backups, recovery, API, email ingestion, monitoring and security. Restore a sample archive and verify search, previews and metadata before importing sensitive production records.
Decision record
Document the tested document types, retrieval results, permission model, exception handling, required edition or modules, migration assumptions and named operational owner. Keep unsupported pricing, customer-size, language and device values blank until a current official source maps them exactly to the BBS field.