win-j logo
Personal Tech Blog

Document Utility

PDF Excel Table Extractor

Extract table data from PDF documents and save it as CSV or Excel files. Analyze text-based PDF tables by page, preview the extracted results, and download the data.

Extract PDF Table Data Analyze text positions inside a PDF, organize them into table-like data, and save the result as CSV or XLSX.
Check Page Candidates Preview extracted table candidates by page before saving only the data you need.
Cleanup Options Use first row as header, remove empty rows, and clean empty columns for Excel-friendly results.
File Not selected
Pages All
Candidates 0
Status Ready

PDF Upload and Extraction Settings

Upload a PDF file, then set the page range and table recognition sensitivity. This tool works best with text-based PDF files.

PDF Upload

Drag & Drop or

Upload 1 PDF / Analyze pages / Extract table candidates / Save as CSV or XLSX

Notice This tool estimates table structure based on PDF text coordinates. Scanned PDFs, image-based PDFs, or documents with complex table lines may produce inaccurate results.

Leave this empty to analyze all pages. Enter a range when extracting only specific pages.

Choose Wide if columns are split too much. Choose Tight if separate columns are merged together.

Choose Wide if rows are split into too many lines. Choose Tight if multiple rows are merged into one.

This file name will be used for CSV and XLSX downloads. The extension is added automatically.

Before Extracting

Accuracy is higher when table cells are aligned with consistent spacing in a text-based PDF.

For documents with multiple mixed tables or complex lines, preview page candidates before saving.

No PDF has been uploaded yet.

Recommended Extraction Presets

Quickly apply page ranges and row/column sensitivity settings based on the PDF table layout.

Extraction Status

Check the selected pages, number of extracted candidates, and current processing status.

Ready
Selected Pages -
Candidates -
Current Status Ready
Upload a PDF and the extraction result will appear here.
Practical guide

Extract PDF table text into CSV or Excel

Read text and its position from a PDF to estimate rows and columns. Preview the result, adjust spacing and export CSV or XLSX. This is text extraction, not OCR or full table-layout recovery.

Prepared & reviewed by
win-j
Last reviewed
2026-09-30

Who this helps

  • People reusing table text from PDF reports
  • People comparing extracted rows with source pages

When to use this tool

  • Move a regular text table into a spreadsheet
  • Export selected pages of a long report

From input to a checked result

  1. 1

    Choose a text-selectable PDF up to 80 MB.

  2. 2

    Leave the range empty or enter pages such as 1-3,5; process at most 200 pages per run.

  3. 3

    Start with row tolerance 6 and column tolerance 16, then adjust the preview.

  4. 4

    Export CSV or XLSX and verify numbers, dates and column alignment.

Example 1

Extract selected pages

Input
Text tables on pages 1, 2 and 4; range 1-2,4
Result
3 page tables and 3 XLSX worksheets

Only pages containing usable text candidates create output tables.

Example 2

Join nearby text columns

Input
Column tolerance 16 changed to 24
Result
Nearby X positions are more likely to become one column

Wider tolerance can also merge separate columns. Compare the preview instead of assuming it is better.

How it works

PDF.js text items are grouped by Y position into rows and X position into columns. Optional empty-row and empty-column removal cleans the result. CSV includes page labels; XLSX contains one worksheet for each extracted page.

Limitations

  • Image-only scans have no text to extract; use OCR instead.
  • Merged cells, rotated text and multiple columns can be grouped incorrectly.
  • Text is exported without reconstructing formulas, formatting or data types.
Your data

The PDF is read and CSV/XLSX is created in this browser. The tool does not upload the file or extracted rows to the win-j server.

Frequently asked questions

Why do visible table borders not fix the columns?

The tool groups text coordinates rather than interpreting table lines.

Does the header option change exported data?

It changes the preview's first-row presentation; the exported rows still contain their original text.

Can I process more than 200 pages?

Use separate page ranges of at most 200 pages per run. Split large documents if memory is limited.