PDF to Excel
Convert a PDF table to Excel in your browser. Columns are recovered from the page layout and figures arrive as real numbers you can total. No upload, no sign-up.
Drop your file here
or click to browse
Use one sheet when the same table continues across pages — the columns are then lined up across all of them.
Writes figures as numbers you can total, keeping currency and percent formatting.
Off by default, and deliberately so: nothing in a PDF says which convention it used, and a date read the wrong way round still sorts and still feeds formulas, so it is wrong in a way nobody notices. Pick the order your document uses and it becomes a real date. Anything that cannot be a valid date that way is left as text rather than forced.
Makes the first row bold and freezes it.
How much blank space counts as a column break. Raise it if one column got split in two; lower it if two columns were merged.
- ProcessingOn your device
- Uploaded to a serverNothing
- InternetNot needed
Don't take our word for it. Turn off your Wi‑Fi, then use this tool. It will work exactly the same, which is only possible because your file never goes anywhere.
Worth knowing
A PDF has no cells — only text at coordinates. Columns are worked out from the strips of blank space that no text crosses, which is reliable on a genuine table and guesswork on a page that merely looks tabular. Always check the result against the original before trusting a total.
Dates stay as text unless you say which way round they are written. 05/12/2024 is two different days depending on which country the document came from, and a wrong guess corrupts a whole column while appearing to have worked — so it is a setting with no default rather than something guessed at. Tell it the order and dates become real dates you can sort and filter.
A scanned PDF contains a picture of a table, not a table. Run it through Searchable PDF first to add a text layer.
Why this is harder than it sounds
A PDF does not contain a table. It contains instructions to draw glyphs at coordinates. When you look at a page and see four neat columns, what is actually stored is several hundred separate pieces of text, each with a position, and nothing anywhere saying which of them belong together. The lines you can see may be drawn rectangles, or they may not exist at all — plenty of tables are held together by nothing but alignment.
So the columns have to be inferred, and every PDF-to-Excel converter is doing the same guessing. The difference is only in how good the guess is and how honest the tool is about making one.
How the columns are found
The obvious method is to group text by its left edge: everything starting at 72 points is column one, everything at 210 points is column two. It works on a page of labels and falls apart on the thing spreadsheets are mostly made of — a column of right-aligned numbers, where 9.50 and 1,204.00 begin at different places entirely. Left-edge clustering scatters that column across three or four phantom ones.
What this uses instead is the whitespace. Every piece of text on the page is projected onto a horizontal line to build a map of which parts of the page width are ever covered by anything. A column boundary is then simply a strip that nothing crosses. It works identically for left-aligned labels and right-aligned figures, because it never looks at where text starts — only at where there is nothing.
The one setting that matters is how wide a strip has to be before it counts as a boundary. Six points is a sensible default. If two columns came out merged, lower it. If one column was split down the middle — which happens when a column of short words leaves a gap in the middle of itself — raise it.
Numbers, and what deliberately is not converted
Figures are written to the sheet as numbers, not as text that looks like numbers. This is the whole point of the exercise: a column you cannot add up is a picture of data. Thousands separators are removed, accountancy-style brackets become negatives, and both European and US decimal conventions are handled. Currency and percentages keep their meaning as well as their value — a cell showing £1,234.56 becomes the number 1234.56 with a currency format, and 12.5% becomes 0.125 formatted as a percentage, so the sum is right and the cell still reads correctly.
Dates are left alone on purpose. 05/12/2024 is the fifth of December in London and the twelfth of May in Chicago, and a PDF carries nothing to say which. Converting them would mean picking one and being silently wrong about half the time, in a way nobody would notice until a report was already out. A text date you have to fix is a nuisance; a date column that is quietly wrong is a real problem.
Checking the result
- Compare the row count in the note against the original. A large mismatch usually means rows were merged, which happens when the source has very tight line spacing.
- Look at the last column. Text that has overflowed into a neighbour tends to show up there first.
- Total a numeric column and check it against any total printed in the PDF. This catches misplaced cells faster than reading every row.
- If a header repeats partway down, the table spans pages — convert again with all pages in one sheet to line the columns up.
If the page turns out to be prose rather than a table, converting it to an editable Word document or pulling the plain text out will get you further than forcing it into columns. There is also a longer walkthrough of getting a table out of a PDF and checking the result.
Nothing is sent anywhere: the PDF is parsed and the spreadsheet written inside this tab. Why that matters for financial documents.
Questions people ask
Why are my dates text instead of dates?
Because nothing in a PDF says whether 05/12/2024 means the fifth of December or the twelfth of May, and a date read the wrong way round still sorts, still filters and still feeds formulas — it is wrong in a way nobody notices. So it is off unless you choose the order. Set 'Read dates as' to day-first or month-first and they become real Excel dates. Anything that cannot be a valid date under the order you chose is left as text rather than forced, so a stray value stands out instead of quietly becoming the wrong day.
Will the columns come out right?
On a normal table, yes. Columns are found by looking for vertical strips of the page that no text ever crosses, which handles right-aligned figures as well as left-aligned labels — the reason simpler tools mangle numeric columns is that they cluster left edges, and a column of right-aligned numbers has a different left edge on every row. Where it struggles is a page with no real column structure: an invoice with a free-form address block, or text in two magazine-style columns. If a column has been split or merged, adjust the column gap setting and convert again.
Can I add up the numbers afterwards?
Yes — that is the point. Figures are written as real numbers rather than text, so SUM works immediately. Currency and percentages keep their presentation too: £1,234.56 becomes the number 1234.56 formatted as currency, and 12.5% becomes 0.125 formatted as a percentage, so the arithmetic is right and the cell still reads the way it did in the PDF.
My table spans several pages. Do I get one sheet or many?
Your choice. One sheet per page is the default and keeps each page separate. Choosing a single sheet works out one set of column positions from every page at once and stacks the rows, which is what you want when the same table continues over a page break — otherwise each page could end up with slightly different columns.
Why is everything in one column?
Because the page has no column structure to find, which almost always means it is prose rather than a table. The note under the result tells you when only one column was detected. For a page of text, PDF to Text or PDF to Word will give you something more useful.
Is my file uploaded?
No. The PDF is read and the .xlsx is built inside your browser, which is why it works with your network switched off. Financial statements and payroll tables are exactly the sort of thing people should not be posting to a stranger's server.
How to use it
How-to guides · 13 August 2026 · 7 min
How to convert a PDF to Excel without retyping it
Get a table out of a PDF and into a spreadsheet you can actually calculate with — in your browser, no upload. Plus how to spot the mistakes a converter makes.
How-to guides · 14 August 2026 · 6 min
Excel to PDF: how to stop it splitting across pages
Turning a spreadsheet into a PDF usually goes wrong in one specific way: the columns spill onto extra pages. Here's why, and the four fixes in the order worth trying.