Treeet
PDFs are great for reading and terrible for working with numbers. If you’ve ever tried to copy a table out of a PDF and paste it into Excel, you know the result: everything lands in one column, numbers come through as text, and rows break in odd places.
This guide shows how to turn a PDF table into a proper spreadsheet, how to clean up the result, and what to do when the PDF turns out to be a scan.
First, check what kind of PDF you have
There are two very different kinds of PDF, and it decides everything that follows.
- Text-based PDFs are created by software: an export from accounting software, a report saved as PDF, an invoice generated by a website. The text is stored as real characters.
- Scanned PDFs are pictures of pages, made by a scanner, a phone camera or a fax. They may look identical, but there’s no actual text inside, only pixels.
To tell them apart, open the PDF and try either of these:
- Select some text in the table by dragging over it. If it highlights word by word, it’s text-based. If you can only draw a box, or nothing happens, it’s probably a scan.
- Search (Ctrl+F or Cmd+F) for a word you can see in the table. If the search finds it, the text is there.
Text-based PDFs convert well. For scans, skip ahead to the section on scanned PDFs.
How to convert a PDF table to Excel
Open the PDF to Excel converter. It runs in your browser, so the PDF stays on your device.
- Add your PDF. You can add several and convert them all at once.
- Choose your settings (explained below).
- Click Convert.
- Download the
.xlsxfile and open it in Excel, Google Sheets, LibreOffice or Numbers.
The settings
Sheets decides how pages are laid out:
- All pages in one sheet (default) stacks every page’s rows into one sheet, with a blank row between pages. Best for one long table that runs across many pages.
- One sheet per page gives each page its own tab, named Page 1, Page 2 and so on. Best when each page holds a different table.
Turn numbers into numbers is on by default. It converts text like “1,234.50” into the number 1234.5 so you can add it up. It also understands a few common formats, such as negatives in brackets like “(250.00)”, amounts with a trailing minus, currency symbols, and European-style numbers like “1.234,50”. Long digit strings such as account numbers stay as text, and so do numbers with leading zeros, so they don’t get mangled.
Join wrapped cell text is on by default too. When a cell’s text wraps onto a second line in the PDF, it rejoins the pieces into one cell instead of creating a half-empty extra row.
Prefer CSV?
If you’re importing the data into another app or a database, CSV is often more convenient. The PDF to CSV converter does the same table extraction and saves plain-text rows.
You can choose the separator: comma, semicolon or tab. If you use Excel with European regional settings, where the comma is the decimal separator, pick semicolon so the file opens in the right columns.
Tips for a cleaner result
Convert only the pages you need
Reports often have a cover page, a summary and notes before the table you care about. Use Extract PDF pages to pull out just the table pages (for example 4-9), then convert that smaller file. You’ll spend less time deleting junk rows.
Check the header rows
Tables with headers split over two lines, or merged header cells, often come out as two header rows. Merge them by hand into one row of column names.
Remove repeated headers and footers
When a table runs over several pages, its header row usually repeats at the top of each page, and page footers may appear in between. In Excel, filter the first column and delete the repeated rows.
Spot-check the totals
Add up a column in Excel and compare it with the total printed in the PDF. If they match, you can trust the conversion. If not, look for a row that was split or merged.
Watch out for dates
Dates are kept as they appear in the PDF. Excel may interpret them differently depending on your regional settings (day/month versus month/day). Check a few dates where the day is greater than 12 to be sure.
What about scanned PDFs?
The converter reads the text stored inside the PDF. A scanned PDF has no text inside, only images, so there’s nothing to extract. You’ll see a message saying the PDF has no selectable text.
Treeet doesn’t do OCR (optical character recognition), the step that “reads” text from an image. You have a few options:
- Ask for the original. If someone sent you the PDF, ask whether they can export the data as Excel or CSV, or send a PDF generated directly from their software. This gives the most accurate result.
- Run OCR first. Some PDF apps and scanner apps can add a searchable text layer to a scanned PDF. Once the PDF has selectable text, try the converter again. OCR can misread characters, especially in numbers, so check the figures carefully.
- Retype it. For a small table, typing it into a spreadsheet may honestly be the quickest and most reliable way.
Password-protected PDFs
If the PDF asks for a password when you open it, the converter can’t read it. If you know the password, use Unlock PDF to save a copy without it, then convert that copy. Unlock PDF only works when you know the password; it can’t remove one you don’t have.
Bank statements are a special case
Bank statements are tables too, but they have quirks: dates printed once per day, descriptions that wrap over two lines, and separate debit and credit columns. The bank statement to CSV converter is built for exactly that. It outputs one clean row per transaction with date, description, amounts and balance. See our guide on turning bank statement PDFs into CSV for details.
Limits worth knowing
PDF tables don’t store “cells” the way a spreadsheet does. They’re just text placed at positions on a page, so the converter has to work out the rows and columns from the layout. That works well for typical reports, invoices and statements. Very complex layouts, such as tables inside tables, rotated text or charts with labels, may need some tidying afterwards.
When a conversion looks off, try the other Sheets option, or switch Join wrapped cell text off and on to see which gives cleaner rows.