How to copy a table from a PDF to Excel without it falling apart
You select a table in a PDF, copy it, paste it into Excel, and get one long column of text, or three columns where there should be six. This guide explains why that happens, shows three ways to get the table across properly, and works through one real example so you can see exactly what you get.
Why a copied table falls apart
A table in a PDF is often not stored as a table. Many PDFs only record where each piece of text sits on the page. The rows, columns and borders you see are the result of that placement. When you copy, your PDF viewer has to guess where one cell ends and the next begins, and Excel then guesses again when you paste. That is why columns merge, wrapped lines turn into extra rows, and numbers arrive as text you cannot add up.
Three ways to do it
1. Excel's own PDF import. Excel can read tables straight from a PDF: on the Data tab choose Get Data, then From File, then From PDF, as Microsoft's support page describes. Microsoft documents the feature as the Power Query PDF connector. It shows what it found in a Navigator window so you can pick what to load. Microsoft's page also notes that rows spanning several lines may not be identified properly and can need cleaning up afterwards. If you already have Excel and a digital PDF, this is a sensible first try.
2. Copy, paste, then split the columns. For a small, simple table this is often enough. Expect to fix the column breaks by hand, and check every number afterwards.
3. Convert the table before it reaches Excel. A converter finds the table, rebuilds its rows and columns, and writes a real workbook. This is what PDF to Excel on this site does: it detects each table and writes an .xlsx file with one sheet per table.
A worked example
To see what really comes out, we tested a one-page PDF with a small sales table that we made ourselves. The data is invented. We chose values that usually go wrong: account codes that start with a zero, three different date styles, thousands separators, a currency and percentages. We uploaded it to PDF to Excel and opened the workbook it produced. Every value in the table matched the PDF. What mattered is what Excel was told each value was:
| In the PDF | In Excel | Why |
|---|---|---|
00417 (account code) |
Text, 00417 |
The leading zero is kept. As a number it would show as 417. |
2025-07-14 |
A real date | The order of year, month and day is unambiguous. |
15 July 2025 |
A real date, shown as 2025-07-15 | The month is written out, so it cannot be misread. |
03/04/2025 |
Text, exactly as written | It could be 3 April or 4 March. A tool cannot know which, so it does not guess. |
1,240 |
The number 1240, shown with a thousands separator | You can sum it. |
$18,450.00 |
The number 18450, shown as a dollar amount | You can sum it, and it still looks like money. |
12.5% |
The number 0.125, shown as 12.5% | Percent cells calculate correctly. |
-3.2% |
The number -0.032, shown as -3.2% | Negative values stay negative. |
The workbook is not a copy of the page's design. Our sample's bold "Total" row came out as an ordinary row, because the workbook styles its own header (bold, with a light fill and a frozen first row) instead of copying the PDF's styling. The content is what carries over. If you are wondering why some of this is lost, the next guide goes through it.
Check the result before you trust it
Any conversion deserves a quick check. Here are three that take under a minute:
- Add up a column and compare it with the printed total. In our sample, the four Units values add up to 4,855 and the four Revenue values add up to $71,941.25, matching the Total row in the PDF.
- Look at the cells left as text. Codes and ambiguous dates are kept as written on purpose. If you know the document is day-first or month-first, convert those dates yourself.
- Count the columns and rows against the original. A missing column is easier to spot than a wrong digit.
When it will not work as well
- Scanned PDFs. A scan is a picture of a page, so the text has to be read with OCR first, and that can misread characters. The result carries a "Review recommended" warning. Compare totals against the original.
- Tables that are not detected. In our own tests with generated sample pages, some clean-looking tables were not detected as tables at all, and the result did not always say so. Treat a missing table, or any warning, as a reason to look at the original page, and compare every table you rely on with the PDF.
- Text outside tables. Only tables are exported. Paragraphs are not part of the workbook.
- Merged cells and unusual layouts. Tables with merged cells, or cells that span several rows, are the hardest case for any tool. Check those results line by line against the original.
For scanned documents, see how to extract text from a scanned PDF.
What happens to your file
PDF to Excel uploads your PDF to our servers, where open-source software processes it. It is not sent to any third-party AI or document API. The uploaded PDF is deleted as soon as processing finishes, and the results stay for 60 minutes so you can download them. Files can be up to 50 MB and 100 pages. If a file must not leave your device, do not upload it. The full details are on the privacy page.