Docyield
Document extraction, not just OCR

The data your documents yield.

Upload an invoice, bank statement, or receipt. Get back clean Excel, CSV, and JSON — every field scored, sourced, and ready to use.

Convert a document freeView API docs
50+ free tools•Excel · CSV · JSON · XML•No signup for your first file
invoice_acme_april.pdfextracting
scanned · skewed · 1 page
supplierAcme Ltd0.96
invoice_noINV-100450.93
date2026-04-120.98
vat€240.500.71
total€1,240.500.98
ibanNL91•••04170.94

Built for the documents businesses run on

InvoicesBank statementsReceiptsPDF tablesPurchase orders

The difference

OCR reads text. Docyield understands the document.

Anyone can pull characters off a page. The value is knowing that this number is the total, that one is the VAT, and how confident to be about each.

Raw OCR output
ACME LTD 14 Mill Rd
INV-10045 12/04/26
Consulting 10 100 1000
VAT 240.50
TOTAL 1240.50 EUR
NL91 ABNA 0417 ...
Docyield output
supplier_name "Acme Ltd"
invoice_number "INV-10045"
total_amount 1240.50 · EUR
tax_amount 240.50 review
line_items [ 1 row ]
iban checksum ✓

Every field carries a value, a confidence score, the page it came from, and the source text it was read from.

More than OCR

Built to automate, not just read

Extract

Turn unstructured documents into labelled fields and line items, not just a wall of OCR text.

Convert

Export the same result as JSON, CSV, Excel, or XML — typed and ready to import.

Classify

Detect the document type automatically, so the right fields are pulled every time.

Validate

Reconciliation checks and confidence signals flag the few documents worth a review.

Free tools

Start with the converter you came here for.

Free to try, no account needed for your first file. Real structured output — not a watermarked preview.

See all tools
Invoice to Excel
PDF / scan → XLSX · CSV · JSON

Supplier, invoice number, dates, VAT, totals, and line items — pulled into clean spreadsheet rows.

Bank statement to CSV
PDF / scan → CSV · XLSX

Every transaction with date, description, amount, and running balance — continuity checked row by row.

Receipt OCR to JSON
Photo / PDF → JSON · CSV

Merchant, date, tax, tip, total, and line items from a crumpled photo — straight into your app.

PDF table extractor
Native / scanned PDF → XLSX · CSV

Detect tables across multiple pages and keep rows and columns intact, even when there's more than one.

Purchase order to Excel
PDF / scan → XLSX · CSV

PO number, buyer, supplier, delivery dates, payment terms, and every line item with quantity and price.

Accounts payable OCR
PDF / scan → CSV · XLSX · JSON

Capture supplier invoices for touchless AP — vendor, PO reference, due date, line items, and totals.

How it works

Upload to clean data in four steps.

Step 1
Upload

Drop in a PDF, scan, or photo. Docyield detects the document type automatically.

Step 2
Extract

Fields are read into a stable schema with a confidence score and source text for each value.

Step 3
Review

Low-confidence fields are flagged for a glance. Validation checks the math before you trust it.

Step 4
Export

Download Excel, CSV, or JSON — or send results straight to your API or webhook.

For developers

Document extraction in one API call.

Upload PDF, scan, or image. Get structured JSON with confidence scores, field evidence, and validation results. Async jobs for big files, webhooks when they finish.

API docs and self-serve keys are not published yet. The free tools above use the same extraction engine.

extract.sh
# Send a document, get structured data back
curl -X POST https://api.docyield.com/v1/extract/invoice \
  -H "Authorization: Bearer $DOCYIELD_KEY" \
  -F "file=@invoice.pdf"

# → response
{
  "document_type": "invoice",
  "fields": {
    "supplier_name": { "value": "Acme Ltd", "confidence": 0.96 },
    "total_amount":  { "value": 1240.50, "currency": "EUR" },
    "tax_amount":    { "value": 240.50, "review": true }
  },
  "line_items": [ { "description": "Consulting", "total": 1000 } ]
}

Frequently asked questions

Do I need to sign up to use the free tools?

No. You can convert your first document without an account. Paid plans add batch processing, API access, webhooks, and higher limits.

What file types can I upload?

PDF, PNG, JPG, WEBP, and other common image formats — including scanned and photographed documents.

How is this different from plain OCR?

OCR reads characters. Docyield understands the document and returns labelled fields against a consistent schema, ready to use without manual mapping.

Is my data kept private?

Uploaded files are used only to produce your result and are not used to train models. Paid plans add controls for sensitive and high-volume workloads.

Get started free

Stop retyping what's already on the page.

Upload your first document and see what it yields — clean, structured, and checked in seconds.

Try the free converter