Copying a table out of a PDF and pasting it into Excel usually gives you one long column of numbers, or every row crammed into a single cell. That's not Excel being difficult. The PDF never contained a table in the first place.
PDF to Excel rebuilds the table from where the text sits on the page and gives you a real spreadsheet: rows, columns, and numbers that Excel treats as numbers. It works best on what most people need it for: bank statements, invoices, price lists and simple reports.
Why a PDF table isn't really a table
A PDF page is a set of drawing instructions. "Put the text 1,250.00 at this position in this font." "Draw a line from here to here." There's no instruction that says "this is row 4, column 3". The table you see is just text that happens to line up, sometimes with lines drawn around it.
So any converter has to guess. The tool looks for two kinds of table:
- Ruled tables, with lines drawn between the cells. These are the easy case. The lines mark the cells exactly, and cells that span several columns or rows become merged cells in Excel.
- Unruled tables, where columns are only implied by text lining up. The tool looks for blocks where at least three rows have text in two or more aligned columns.
It also tries not to see tables where there aren't any. Bulleted lists and pages set in two columns of running text are deliberately left as text. And when a cell's text wraps onto a second line, that line is joined back onto the row above instead of becoming a row of its own, which is one of the more common messes in converted tables.
Converting a PDF on EddiesPDF
The conversion runs in your browser. The PDF isn't uploaded, which matters when it's a bank statement.
- Open PDF to Excel and drop your PDF on the page.
- Decide whether to tick "All tables in a single spreadsheet". Off (the default) gives one sheet per page, named Page 1, Page 2 and so on. On puts everything on one sheet, one table below the other.
- Optionally, open "More options". Set "Content" to "Tables only" to drop headings and paragraphs, and type a range into "Pages", like 2-5, if you only need part of the document.
- Click "Convert to Excel" for an .xlsx file, or "Convert to CSV" for plain text.
- Download and open the file.
For a multi-page statement you want to sort or total, I'd tick the single-sheet option. For an annual report where each page has a different table, one sheet per page is easier to find your way around.
If you pick "Tables only" and nothing on the pages looks like a table, you still get a file: the text is placed line by line instead, and the results screen tells you so.
Excel or CSV
Excel keeps more: number formats, merged cells and multiple sheets in one file. CSV is plain text that any spreadsheet, database or script can read, but it has no formatting and only one sheet per file, so a multi-page conversion without the single-sheet option downloads as a ZIP of CSV files. Unless you're feeding the data into something else, choose Excel.
Scanned PDFs need OCR first
If the tool says "No text was found in this PDF", you almost certainly have a scan. A scanned page is a photo of paper. There's no text on it to position, so there's nothing to build a table from.
The fix is to run it through OCR PDF first. That adds an invisible layer of recognised text over each page image, and PDF to Excel can then read that layer like any other text. The guide to making a scanned PDF searchable walks through the settings.
Be realistic about the result. OCR reads characters from pixels, and on a slightly blurry scan it will sometimes read 8 as 3, or 0 as O. In a paragraph that's a typo. In a column of amounts it's a wrong total. Straighten crooked scans with Deskew PDF before OCR, and check the numbers afterwards against the original.
Cleaning up the spreadsheet
The tool already does some of this for you. Text that looks like a number becomes a real number: "1,234.50", "(300)" as a negative, "12%" as a percentage, and European style "1.234,56". Currency symbols like $, €, £ and ₹ are stripped so the value can be summed. Values with leading zeros, like an account number "00123", stay as text so the zeros survive.
What's left for you is usually one of these:
- Repeated header rows. On a single sheet, a statement that repeats "Date, Description, Amount" at the top of each page will have that row several times. Turn on a filter (Data, Filter in Excel), filter the first column for the header text and delete those rows.
- Dates stay as text. The tool doesn't convert dates, because "03/04/2026" means March in the US and April almost everywhere else. Select the column and use Data, Text to Columns, choose Date and the right order (MDY or DMY) in the last step.
- Ambiguous commas. "1,234" is read as one thousand two hundred and thirty-four. If the document uses a comma as the decimal separator and has values like that, check them.
- Columns split or merged. When columns sit very close together, two can end up in one cell, or one wide column can split in two. Fix by hand, or use Text to Columns on the merged one.
Check the totals before you trust it
The one habit worth building: find a number the PDF already totals for you (a closing balance, an invoice total, a column sum at the bottom of a page) and make Excel compute the same thing with SUM. If the two match, the conversion is almost certainly right. If they don't, compare the difference with the rows: an amount that matches one row exactly usually means that row was skipped or doubled, and a difference like 5.00 suggests an 8 that OCR read as a 3. It takes a minute and it's the difference between a spreadsheet you can rely on and one that looks right.
Also useful: OCR PDF · PDF to text
