Pricing

Pay per usage

Try for free

Go to Apify Store

Document to Structured Data — PDF, Excel, Word, CSV, Scans

Try for free

Turns PDFs, scanned images, XLSX, CSV and DOCX files into clean structured JSON: text blocks with page numbers, tables as lists of lists, document metadata and a flat CSV-friendly row. One dataset item per document, at $3.00 per 1,000 documents.

Pricing

Pay per usage

Rating

0.0

(0)

Developer

Spudnicks

Actor stats

Bookmarked

Total users

Monthly active users

2 days ago

Last modified

Document to Structured Data

Turns documents into clean, structured JSON. Send PDFs, scanned images, Excel workbooks, CSVs or Word files; get back one dataset item per document — text broken into blocks with page numbers, tables as real rows and cells, document metadata, and a flat record that exports straight to CSV.

Built for the ordinary case that wastes the most time: a folder of documents that someone has to open, read and retype.

What you get per document

Field	Meaning
`source`, `fileName`, `docType`	where it came from and what it is
`pageCount`, `sheetCount`, `unitCount`	pages or sheets, whichever applies
`blockCount`, `charCount`	size of the extracted text
`blocks[]`	text blocks with page number, type, text and position
`tables[]`	every table as rows and cells, not a picture of a table
`metadata{}`	producer, author, dates, page size, sheet names, OCR engine
`text`	the whole document as plain text with `[page n]` markers
`warnings[]`	anything that degraded, so nothing fails silently
`flat{}`	scalar-only view for spreadsheet and CSV export

Input

{
  "documentUrls": ["https://example.com/report.pdf"],
  "maxPages": 50,
  "ocrImages": true,
  "includeFlatRow": true
}

Point it at public URLs. Mixed batches are fine — a PDF, a workbook and a scan in one run all come back in the same shape.

Output

One dataset item per document, for example:

{
  "fileName": "quarterly.pdf",
  "docType": "pdf",
  "pageCount": 12,
  "tableCount": 3,
  "tableCellCount": 148,
  "tables": [{ "page": 2, "rowCount": 4, "columnCount": 3, "rows": [["Region", "Q1", "Q2"], ["North", "1240", "1380"]] }],
  "warnings": []
}

Pricing

Pay per event: $3.00 per 1,000 documents processed. Failed downloads and unsupported files are not charged.

Honest limits

Scanned pages go through OCR; handwriting, stamps and very low-resolution scans produce text that needs checking. Those cases are reported in warnings[], not hidden.
A layout that mixes columns and tables can split a table into two blocks; the warnings list says when it happened.
Password-protected files are reported as an error rather than guessed at.

Support

Open an issue on this actor's page. Include the warnings[] from your run — that is usually enough to see what the document did.

Real-Time Lens Data API

openweb-ninja/real-time-lens-data

Fast and reliable Google Lens API - search by image and get visual matches, exact matches, object detection, and image to text (OCR) in real-time.

OpenWeb Ninja

OCR & Document Extractor – PDF & Image to Text, JSON, Word

lofomachines/ocr-document-extractor

Convert scanned PDFs and images into clean, structured text in bulk. Export to JSON, Markdown, DOCX, TXT or HTML with tables and layout preserved.

Lofomachines

AI Document Assistant

devninja/ai-document-assistant

This actor analyzes uploaded documents using AI to extract and process information. It helps businesses quickly get answers from their documents and automate decision-making.

Devinja

5.0

(2)

Universal Document Format Transformer

actorify/universal-document-format-transformer

Universal Document Format Transformer: a cloud-based Apify Actor that converts documents (PDF, DOCX, PPTX, HTML, TXT) into Markdown, JSON, CSV, HTML or TXT using Pandoc. Easy REST API for automations (n8n, Zapier, Make), production-ready error handling, and security controls.

fanio zilla

Website JSON-LD Structured Data Scraper

cliqtomedia/website-jsonld-structured-data-scraper

Extract JSON-LD objects from public HTML URLs. Keep schema types, original values, script indexes and graph paths. Get clear invalid JSON and page error results.

Cliqto Media

Structured Extract

gastronomic_desk/structured-extract

Only pay when it works. $0.05 per verified extraction — nothing charged on failure or retries. Extract structured JSON from any webpage using your own schema. AJV-validated output guaranteed. Compatible with Groq, OpenAI, Together AI, and Ollama.

Herbert Yeboah

Pandoc Document Converter - HTML to Markdown, DOCX, EPUB, PPTX

scrapeworks/pandoc-document-converter

Convert documents between formats with Pandoc in the cloud: HTML to Markdown for LLMs and RAG, Markdown to Word DOCX, EPUB e-books, PowerPoint PPTX, LaTeX, reStructuredText and more. Feed it URLs or raw text, get one converted document per input.

Nicolas van Arkens

Document to Markdown – PDF, Word, PowerPoint, Excel + OCR

kris_fili/document-to-markdown

Convert PDF, DOCX, PPTX, XLSX, HTML, EPUB and images to clean Markdown or plain text for LLMs and RAG. Tables kept, scanned pages read with OCR in 100+ languages, optional RAG chunks with page numbers.

Krzysztof Filipiak

Document to Markdown for AI — Word, PowerPoint, Excel, PDF

tidydata/document-to-markdown

Convert Word (DOCX), PowerPoint, Excel, EPUB, HTML, CSV, JSON and text-based PDF files to clean Markdown for LLMs and RAG: PDF to Markdown, DOCX to Markdown and more. Optional chunking with overlap. Tables kept. $3 per 1,000 documents; failed files are free.

tidydata

Document → Structured Data Extractor

warant/document-structured-data-extractor

PDF or image documents in — full text, detected tables and business fields (dates, amounts, totals, reference numbers, emails, phones) out as JSON/CSV. Deterministic parsing with Tesseract OCR for scans. No LLM, no login.