Free PDF to Excel Converter
Extract tables from text-based PDFs into Excel XLSX or CSV. Rows and columns are rebuilt from text positions, with a preview and number options.
Loading tool…
How to extract tables from a PDF into Excel or CSV
Statements, price lists and reports exported from software contain real text with exact positions. This tool groups that text into rows by height on the page and into columns by horizontal position, then writes a spreadsheet.
- Choose a text-based PDF up to 50 MB. Enter its password only if the PDF asks for one.
- Optionally enter a page range such as 2-5, and choose one sheet per page or all pages in one sheet.
- Choose how numbers are written in the document (1,234.56 or 1.234,56) or keep everything as text.
- Select Extract tables and review the preview. If columns are split or merged wrongly, move the sensitivity slider and extract again.
- Download Excel (.xlsx) or CSV, or copy one page to paste into any spreadsheet.
When PDF to Excel conversion helps
Banks, accounting systems, suppliers and governments often publish data only as PDFs. Retyping a statement or a price list is slow and error-prone. If the PDF was created by software rather than scanned, its text is stored with exact positions, and those positions can be turned back into rows and columns.
Typical uses are importing card or bank statements into a budget, comparing supplier price lists, analysing published statistics tables and moving an invoice's line items into a spreadsheet.
This is different from PDF to Word, which rebuilds flowing paragraphs. Use PDF to Excel when the content is a grid of values; use PDF to Word when it is mainly prose.
Worked example: a city table with grouped thousands
Our test PDF contains a table with three columns: City, Country and Population, where the population figures are right aligned. The rows include Lagos, Nigeria, 15,388,000; São Paulo, Brazil, 12,325,232; and Tokyo, Japan, 14,094,034. The extractor rebuilt exactly four rows and three columns, accents included.
With number conversion set to the 1,234.56 style, 15,388,000 is written to Excel as the number 15388000, so it can be summed and sorted. With the 1.234,56 style, used in much of Europe and Latin America, the text 1.234,56 becomes 1234.56. Accounting negatives such as (1,250.00) become −1250.
Values that only look numeric are kept as text: the code 007 keeps its leading zeros, 12% stays as written, and dates are not reinterpreted. That way nothing changes silently, and you can convert those columns deliberately in your spreadsheet.
How rows and columns are rebuilt
pdf.js reads every piece of text on a page with its position and font size. Pieces whose baselines sit within half a text height of each other are grouped into the same row, working from the top of the page down.
Within a row, pieces that are closer together than the column split sensitivity (by default one text height) are joined into one cell, so a phrase like 'Wireless keyboard' stays together while a wide gap starts a new cell. The tool then looks at the horizontal extent of every cell in rows that have two or more cells and merges overlapping ranges into column bands. Titles that span the page are ignored when the bands are worked out, so they do not merge every column into one.
Each cell is placed in the band it overlaps most, which keeps right-aligned numbers and left-aligned text in the right columns. The workbook is written with SheetJS, with column widths set from the longest value.
Common mistakes and fixes
Using a scanned PDF. If the preview is empty or pages are listed as having no text, the page is an image. Export it with PDF to JPG and run Image to Text first, then paste the text into a spreadsheet.
Picking the wrong number style. If 1.500 means one and a half in your document, choose the comma decimal style; otherwise it would be read as one thousand five hundred. When unsure, choose Keep as text.
Expecting merged cells or wrapped cells to survive. A cell whose text wraps onto two lines in the PDF becomes two rows. Fix these in the spreadsheet, or adjust the sensitivity slider and extract again.
Useful tips and limitations
- Scanned PDFs are images with no text layer. Export the pages with PDF to JPG and run Image to Text (OCR) first.
- Codes with leading zeros, percentages and dates stay as text so nothing is silently changed. Convert them in your spreadsheet if needed.
- Extract only the pages that hold the table (for example 3-7) to keep cover pages and terms out of your workbook.
- One sheet per page makes it easy to check each page against the PDF; switch to one sheet when you want a single list to sort and filter.
Common questions
Why are some PDF pages empty in the result?
Those pages have no selectable text, usually because they are scans or photos. They need OCR (text recognition) before a table can be extracted.
Why did two columns merge into one?
The columns sit very close together. Lower the column split sensitivity and extract again, so smaller gaps start a new cell.
Is the PDF uploaded to a server?
No. Text is read with pdf.js in your browser and the spreadsheet is created locally.
Can it convert a password-protected PDF?
Yes, if you know the password. Enter it before extracting. It is used only in your browser.
Why does the CSV contain all pages?
CSV files cannot hold several sheets, so the CSV always combines every extracted page into one table, separated by blank rows.
Will Excel show accented characters correctly in the CSV?
Yes. The CSV is saved as UTF-8 with a byte-order mark, which tells Excel to read accents and other non-English letters correctly.
Your files are processed in this browser and are not uploaded to ToolWorld Online. Nothing is kept after you close the tab, so save the finished download first.