Heyyel Data ExtractorHeyyel Data Extractor

How Heyyel Data Extractor works

From document to structured data in a few steps

Upload your document, choose its type, configure the fields and language, then let Heyyel Data Extractor extract structured data you can review and export.

Extract data now → See how it works
77 languages Custom fields OCR XLS · CSV · JSON 1 credit per page
document.pdf → structured data
PDF
invoice_august.pdf Uploaded document · 1 page
↓ Invoice preset · Auto-detect language
SupplierText
Invoice dateDate
TotalNumber
Line itemsArray
↓ Run extraction
{
  "supplier": "Example Company",
  "invoice_date": "2026-08-18",
  "total": 1245.90
}

The idea

Your data is already in the document

Invoices, contracts, bank statements, receipts and other documents already contain the information your team needs. The difficult part is turning that information into data that can be used in spreadsheets, software, databases and automated workflows.

Heyyel Data Extractor lets you define what you need first, then focuses the extraction on those fields instead of forcing you to search through an entire document manually. For teams still relying on manual data entry, this means fewer repetitive copy-and-paste steps between documents and business systems.

No manual copy-pasteReduce the need to open documents and re-enter values one by one.
No fixed outputChoose the fields and structure that match your workflow.
Scans are supportedUse OCR when documents are image-based or scanned.
Structured resultsReview selected information and export it to XLS, CSV or JSON.
01Upload
your document
02Choose a
document type
03Configure fields
& language
04Run
extraction
05Review fields
& preview
06Export XLS,
CSV or JSON

Step by step

How it works

The same workflow can be used for standard business documents and custom document types.

01

Upload your document

Start by uploading the file you want to process. Heyyel Data Extractor works with digital documents, scanned files and document images. If your workflow starts with PDFs and you only need selected values rather than the full document, the same process can be used to extract specific data from a PDF.

  • PDF documents
  • PNG and JPG / JPEG images
  • WebP and TIFF files
  • Digital or scanned content
Upload documentStep 1
↑
Drag & drop your document or choose a file from your device
PDFPNGJPGWebPTIFF
02

Choose your document type

Select the type of document you are processing. Heyyel Data Extractor can start with suggested fields for common document types, or you can use Custom / Blank extraction when you want to build the structure yourself.

  • Invoice and Receipt
  • ID Card / Passport and Business Card
  • Contract and Bank Statement
  • Custom / Blank for your own document structure
Choose document typeStep 2
InvoiceSuggested invoice fields
ReceiptMerchant, date, totals...
ID Card / PassportIdentity document fields
Business CardContact information
ContractParties, dates, terms...
Bank StatementBalances and transactions
Custom / BlankBuild your own schema
03

Configure fields and document language

Review the suggested fields or build your own. For each field, you can define how the extracted value should be structured and add an extraction hint when extra context is useful. This is the same flexibility used in custom data extraction, where the output schema is defined around your own document and workflow. Then choose the document language or let Heyyel Data Extractor detect it automatically.

  • Edit the field label and key
  • Choose the data type and mark fields as required when needed
  • Add an extraction hint or instruction
  • Use language auto-detection or select the language manually
Fields to extractEditable
LabelInvoice number
Keyinvoice_number
TypeText
RequiredYes
Extraction hint: Extract the invoice reference or invoice number shown on the document.
+ Add field
Document language(s)Choose manually or detect automatically
Auto-detect
04

Run the extraction

When your document type, fields and language settings are ready, start the extraction. Heyyel Data Extractor reads the document and identifies the values that match the fields you configured. For scanned or image-based files, OCR is used when text recognition is required; OCR and data extraction work together here, but they handle different parts of the process.

  • Start processing with your configured extraction schema
  • OCR when text recognition is required
  • Extract the values requested in your fields
Run extractionProcessing
→
SupplierExample Company
Invoice date18 Aug 2026
Total€1,245.90
05

Review the extracted data

Review the extracted fields next to the original document preview. This makes it easier to check the returned values while keeping the source document visible.

  • Review extracted fields and values
  • Keep the original document visible alongside the result
  • Check lists, line items and other structured values
Extraction resultReady
Supplier nameExample Company Ltd.✓
Invoice numberINV-2026-00428✓
Invoice date18 Aug 2026✓
CurrencyEUR✓
Total1,245.90✓
06

Export structured data

Once extracted, the information can be used outside the original document. Export structured data for spreadsheets, software, databases and automated workflows.

  • XLS for spreadsheet-based workflows
  • CSV for tabular data and bulk processing
  • JSON for applications, APIs and databases
XLSCSVJSON
{
  "invoice_number": "INV-2026-00428",
  "invoice_date": "2026-08-18",
  "supplier": "Example Company Ltd.",
  "currency": "EUR",
  "total": 1245.90
}

Flexible schemas

Extract simple values or complex structures

Custom fields let you shape the output around your document and your workflow, which is especially useful when layouts vary and AI document data extraction needs to identify the same information across differently structured files.

TXT

Text

Names, clauses, references, descriptions and other written information.

123

Numbers

Amounts, quantities, percentages and other numerical values.

DATE

Dates

Invoice dates, due dates, effective dates and expiration dates.

[ ]

Arrays & objects

Line items, transaction lists and nested document structures.

OCR vs data extraction

OCR reads the document. Data extraction finds what matters.

The two capabilities work together, but they solve different parts of the process.

OCR

Read the text

“What text is inside this document?”

OCR makes scanned or image-based text machine-readable.

Data extraction

Structure the information you need

“Which parts of this information does my workflow need?”

Heyyel Data Extractor identifies your selected fields and returns them in a structured form.

One workflow, many documents

Use the same process across document types

Start from a common document type, extract raw data from the full document, or create a custom structure when your document does not fit a standard template.

📄

Raw data

Everything in the document — no field setup.

Use the full document output — Raw Data Extraction →

The complete workflow

From upload to usable data

The document can stay complex. Your output does not have to be. Define the information you need and turn it into a clean structure that is easier to use elsewhere.

77 languages Custom fields XLS · CSV · JSON 1 credit per page
01
Upload your documentPDF, PNG, JPG, WebP or TIFF
↓
02
Choose a document typeUse a preset or start with Custom / Blank
↓
03
Configure fields & languageLabels, keys, types, required fields, hints and language
↓
04
Run extractionProcess the document with your configured schema
↓
05
Review the resultExtracted fields beside the document preview
↓
06
Export your dataXLS, CSV or JSON

FAQ

How Heyyel Data Extractor works

Upload a document, choose a document type, configure the fields and document language, run the extraction, review the extracted values alongside the document preview, and export the result to XLS, CSV or JSON.

Yes. OCR is used when needed to read text from scanned documents and images. The recognized information can then be processed into the structured fields you requested.

Yes. Choose a preset document type to start with suggested fields, or use Custom / Blank. You can edit field labels, keys and data types, mark fields as required, and add extraction hints.

Heyyel Data Extractor supports common document and image formats including PDF, PNG, JPG, WebP and TIFF.

Extracted data can be exported to XLS, CSV or JSON.

Heyyel Data Extractor uses a page-based credit model: one processed page uses one credit.

Try it on your document

Choose your fields. Upload your document. Get structured data.

Turn the information inside PDFs, scans and images into data you can review, export and use in your workflow.

Extract data now →
77 languages Custom fields OCR XLS · CSV · JSON 1 credit per page