Why PDF Tables Break in Excel (and How to Fix Them)
What a PDF table really is, why columns drift and numbers turn into text, and a short routine that gets a bank statement or invoice into Excel in a form you can sum.
PDF tables break in Excel because a PDF has no table object. It stores text at coordinates, and whatever reads it has to guess where rows and columns are, then guess again whether "1.234,56" is a number. A converter that groups text by position and parses amounts gets you clean columns. You can try that on your own file with PDF to Excel: it runs in your browser, with no upload, no cost and no signup.
The same limits explain odd Word results, so the last section covers PDF to Word too.
What is a PDF table, really?
A set of text fragments, each with an x and y position, sometimes with lines drawn on top. Nothing marks which fragments share a row or whether one cell spans two columns. Only tagged PDFs, made for accessibility, can carry that structure, and plenty of statements and invoices don't.
Whatever extracts the table has to rebuild it from geometry. It groups fragments that sit on the same line, then looks for vertical gaps that no row crosses. That works well on a regular grid and fails in predictable places.
Where columns go wrong
- Two columns that sit very close together get read as one.
- A description that wraps onto a second line looks like a new row.
- A "Totals" label that spans two columns hides the gap between them.
- Rows below the table, like subtotal, tax and total, get glued to the last data row.
That last one is something we hit while testing. With a sample invoice, a general-purpose PDF reader merged "Subtotal, IVA 16%, Total" into the last line item. Reading by position put each total on its own row, so the tool now compares both readings and keeps the cleaner one.
Why do numbers turn into text?
Because the PDF only knows characters. The string "1.234,56" is nine characters, not one thousand two hundred thirty-four and fifty-six hundredths. Excel can't sum it, and its locale makes things worse: a document written with decimal commas pasted into a decimal-point Excel is read as text or as the wrong number.
A good conversion looks at every number in the document before deciding. If it sees 1.234,56 and 12,5, the comma is the decimal; if it sees 1,234.56 and 10.25, the dot is. It also handles parentheses for negatives, a trailing minus, currency symbols and percentages, and it turns dates into real Excel dates. It should leave identifiers alone: an 18-digit account number or a code with leading zeros stays text, or Excel will round or truncate it.
How do I get a PDF table into Excel with clean columns?
- Check that the PDF has selectable text. If you can't highlight a word, it's a scan.
- Open PDF to Excel, drop the file and click "Find tables".
- Choose one sheet per table or everything on one sheet. Leave "Join tables split across pages" on if a statement repeats its header on every page.
- Check the preview: numbers should be right-aligned. If 1.234 appears where you expected 1,234, switch the number format.
- Download the .xlsx, sum a column and compare it with the total printed in the PDF.
Excel 365 also has Data > Get Data > From File > From PDF, which tends to work best on clearly defined tables. If a PDF has no borders or you're on another edition, a browser tool is the quickest alternative.
What about scanned tables?
A scan is a picture, so nothing here can read it until OCR has turned the picture into text. Both tools tell you when a file is a scan instead of handing you an empty file. For plain text from a scan, see how to extract text from a scanned PDF.
And PDF to Word?
Word has the same problem with paragraphs that Excel has with tables. A PDF paragraph is a stack of separate lines, so a converter rebuilds paragraphs, headings, lists and tables from position and font size. Expect editable content, not an exact copy of the layout: columns, fonts and floating boxes can change, and now and then a running header or footer survives as a stray line of text. PDF to Word keeps headings as real heading styles, so the navigation pane works. If you only need the text for an AI chat, read why PDFs paste badly into AI chats.
Frequently asked questions
Why does everything land in column A when I paste a PDF table into Excel?
Because copying from a PDF gives you lines of text, not cells. Each line becomes one row and the spaces between columns are just characters. Excel's Text to Columns can split the lines at fixed widths or delimiters, but a tool that detects the table from the position of the text skips that step and fills the columns directly.
Why do my numbers show up as text after converting a PDF to Excel?
A PDF stores characters, not values. An amount like 1.234,56 or (450.00) is pasted as a string unless the converter understands thousands and decimal separators, parentheses for negatives and currency symbols. The sign that it happened: numbers aligned to the left and a SUM that returns zero.
Does Excel have a built-in way to import a PDF table?
Yes. Recent Microsoft 365 versions have Data > Get Data > From File > From PDF, which uses Power Query. It tends to work best on clearly defined tables, and it isn't available in every edition. A browser tool is a good fallback when you can't use it or when the table has no borders.
Can I convert a scanned PDF table to Excel?
Only after OCR, which reads the letters in the image and turns them into text. To check, try selecting a word in the PDF: if you can't, it's a scan. Docuboxer's PDF to Excel and PDF to Word don't run OCR; when a file is a scan they say so rather than hand you an empty spreadsheet.
How can I tell the conversion got the numbers right?
Sum a column in the spreadsheet and compare it with the total printed in the PDF. If a bank statement's debits, credits and closing balance match to the cent, the columns didn't shift. It takes ten seconds and catches nearly every extraction error.
Is it safe to convert a bank statement or invoice online?
Docuboxer's converters run in your browser and the file isn't uploaded to any server; you can confirm it in your browser's Network tab. With any other converter, check whether it says where the file is processed before uploading financial documents.
Try it on your own PDF
Tables to Excel with real numbers, or an editable Word file. In your browser, free, no signup.
Open PDF to Excel →Related tools
- PDF to Excel: tables to cells, with real numbers and dates.
- PDF to Word: editable content with headings, lists and tables.
- PDF to Markdown: the document structure as clean text.
- Extract images from PDF: the original images in a ZIP.
You might also like: how to convert CSV to Excel without losing formatting.