What DokuScan can do.
No fixed document types. No rigid fields. You define what gets extracted.
Start for freePDF and image recognition
Text-based PDFs are read directly. Scanned documents and photos via OCR.
Text PDFs — no OCR needed
Digital PDFs are processed directly via text extraction — fast and without OCR errors.
OCR for scans and photos
Scanned documents and photos are processed with Google Vision OCR. Even poorly legible originals.
Multilingual
German, English, Polish, Spanish and French are all supported.
Custom templates
Create a template once — reuse it as often as you like.
Define your own fields
You specify which fields you need. The AI extracts exactly those fields from the document.
Three extraction modes
Single fields for forms, rows for tables, or both combined for timesheets and delivery notes.
Reusable
Templates are saved and suggested automatically on your next upload.
Version history
Every change to a template is automatically saved as a version. View earlier states at any time and restore them with one click.
Tables with custom columns
New: read invoice line items, order lists or delivery notes with multiple values per row as a structured table.
Define your own columns
You decide what columns the table has — item, quantity, price, or whatever your document contains.
Every row individually
Each table row is recognized and exported as its own record — not as running text.
Combine with single fields
Capture header data like date or customer together with table rows in a single template.
Submit documents by email
New: every template gets its own email address — send a document as an attachment, DokuScan processes it automatically.
Dedicated address per template
Every template gets a unique email address — ideal for forwarding from scanners or other inboxes.
No login required
Just send an email with an attachment — processing starts automatically, with no manual upload.
Available from the Starter package
Inbox automation is available in every package — not just the expensive tiers.
Export
Structured export — with the column names you defined.
Excel (.xlsx)
Open directly in Excel — columns match your field definitions.
CSV
Semicolon, comma or tab — you choose the delimiter.
JSON
For developers and API integrations.
Batch processing
Upload up to 25 documents at once.
Multi-upload
Upload up to 25 files at once — PDF, JPG, PNG up to 20 MB.
PDF splitting
Multi-page PDFs are automatically split. Each page becomes its own scan.
Duplicate protection
Files that have already been uploaded are automatically detected and skipped.
Review & correct
Check and correct extracted data before exporting.
Inline editing
All extracted fields can be edited directly — individual values and table rows.
Confidence display
Every scan shows how confident the extraction was. Low confidence means extra attention is worthwhile.
Document preview
The original document sits right next to the extracted fields — no switching back and forth.
Automatic validation
New: define rules per field — DokuScan automatically checks every scan and flags issues.
Min/max for numbers
Set value ranges for number fields — for example realistic hour counts or invoice amounts.
Allowed options for text
Define fixed values, e.g. departments or status — deviations are detected immediately.
Date not in the future
Ideal for timesheets and receipts — typos in dates stand out immediately.
Privacy & security
GDPR-compliant. German servers. Your data is never used for training.
German servers
All data is stored in the EU — S3 Frankfurt.
No AI training
Your documents are never used to train AI models.
GDPR-compliant
Complete data separation between accounts. Right to deletion at any time.