PDF Table Extractor: PDF Tables to Excel, CSV & JSON (OCR)

What does PDF Table Extractor do?

PDF Table Extractor finds every table in your PDFs and documents and gives it back as an Excel workbook (.xlsx, one sheet per table), CSV files and structured JSON (header + rows). It works with native and scanned PDFs (OCR), Word (DOCX), PowerPoint (PPTX), Excel, HTML pages and images.

Paste document URLs or Google Drive / Dropbox / OneDrive links and download the tables. No API key, no templates to draw, and you pay only per document processed, however many tables it has.

It runs on the Apify platform, so you also get an API, scheduling, integrations (Google Sheets, Make, Zapier, n8n) and access for AI agents through the Apify MCP server.

Why use it?

• 📊 PDF to Excel without copy-paste: financial statements, annual reports, price lists, invoices, government statistics.
• 🔗 Multi-page tables with the header repeated on each page are merged into one table.
• 🎓 Academic papers: tables drawn with horizontal rules only (LaTeX booktabs), which most tools return as jumbled text.
• 🖨️ Scanned documents: pages without a text layer are read with OCR in 7 languages.
• 🌐 Whole report pages: point it at a web page and enable Also process documents linked from web pages to extract the tables of every PDF linked there.
• 🔢 Numbers stay numbers in Excel (1,234.5 becomes 1234.5), while codes with leading zeros stay text.

How to extract tables from a PDF

1. Click Try for free.
2. Paste one or more PDF (or DOCX/PPTX/XLSX/HTML) URLs.
3. Click Start.
4. Open the files.xlsx link in the results to download the Excel workbook, or export the Tables view as CSV/Excel.

Input

Field	Description	Default
sources	Document URLs (PDF, DOCX, PPTX, XLSX, HTML, images; share links work)	required
saveFiles	Save an Excel workbook and one CSV per table, with download links	true
tableDetection	lines (ruled tables, precise) or text (also borderless tables, experimental)	lines
ocr	auto (only pages without text), always or never	auto
pageRange	e.g. 1-5, 8	all pages
followDocumentLinks	Extract tables from documents linked on a web page	false

{
  "sources": [{ "url": "https://example.com/annual-report.pdf" }],
  "tableDetection": "lines"
}

Output

One row per document. The Tables view shows one row per table. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

{
  "source": "https://example.com/annual-report.pdf",
  "status": "ok",
  "fileType": "pdf",
  "tableCount": 2,
  "pagesProcessed": 12,
  "tables": [
    {
      "index": 0,
      "page": 4,
      "header": ["Year", "Revenue", "Net income"],
      "rows": [["2024", "1,234.5", "210.3"], ["2025", "1,410.2", "251.9"]],
      "csv": "Year,Revenue,Net income\n2024,\"1,234.5\",210.3\n2025,\"1,410.2\",251.9\n"
    }
  ],
  "files": {
    "xlsx": "https://api.apify.com/v2/key-value-stores/.../records/001-annual-report-tables.xlsx",
    "tables": ["https://api.apify.com/v2/key-value-stores/.../records/001-annual-report-table-1.csv"]
  }
}

Data fields

Field	Description
tableCount	Number of tables found
tables[].page / tables[].index	Where the table is
tables[].header / tables[].rows	Table cells as text
tables[].csv	The table as CSV
files.xlsx	Excel workbook with every table of the document
files.tables	One CSV file per table
pagesProcessed, ocrPages	Pages processed and pages read with OCR

How much does PDF table extraction cost?

Event	Price
Run start (per GB of memory)	$0.001
Document processed (any number of pages and tables)	$0.003
OCR page (only scanned pages without text)	$0.01

100 native PDFs cost about $0.30. Failed documents are free. Set Max cost per run and the Actor stops cleanly at that limit.

Use it with AI agents (MCP)

Add https://mcp.apify.com?tools=fguiraud/pdf-table-extractor to Claude, Cursor or any MCP client and ask: "Get the income statement table from this annual report as CSV."

Agent-friendly input: common field names such as url, urls, startUrls, query or keywords are accepted, and an empty input runs a small sample from the input form (with a note in the results) instead of failing.

Need the full text too?

For the whole document as LLM-ready Markdown, RAG chunks and AI field extraction (invoices, contracts), use PDF to Markdown Extractor & Document Parser, built on the same engine.

Only need the plain text and metadata? PDF Text Extractor returns the full text, the text of each page, page count, title and author.

FAQ and limitations

• My table was not found. Try tableDetection: "text" for tables without borders or rules.
• Charts: values drawn in a PDF chart image are not extracted, but PowerPoint chart data is returned as a table.
• Encrypted PDFs: enter the password in PDF password.
• Found a problem or need a feature? Open an issue on the Issues tab. Replies within 48 hours.