How to convert a scanned PDF to Excel

Convert PDF to Excel · 4 min read

You have a scanned or photographed PDF, maybe an old invoice or a printed report, and you need its table in Excel. The catch: there is no text to copy, only a picture of one. Here is how to turn that image into real, editable cells in one click.

A normal PDF has selectable text under the page, so a converter can read the table directly. A scanned PDF does not: it is an image, whether it came from a scanner, a phone photo or a fax. Try to select the numbers and nothing highlights. To get that data into a spreadsheet you first need OCR, optical character recognition, which looks at the picture and recognises the characters, then a tool that rebuilds those characters into rows and columns.

The fastest way: a one-click Chrome extension with OCR

Convert PDF to Excel is a free Chrome extension that handles both kinds of PDF. For a scanned file it uses an optional AI mode to read the image, then exports the table as editable Excel or CSV.

1. Add the extension to Chrome and open your scanned PDF.
2. Click the extension icon and turn on the AI mode for image-only pages.
3. It runs OCR on the scan, recognises the numbers and text, and rebuilds the table column by column.
4. Export to Excel (.xlsx) or CSV and download an editable file.

Why a plain converter fails on a scan

Most PDF-to-Excel tools only read the text layer. Point them at a scan and they return an empty sheet or a single garbled cell, because there is no text layer to read. The image has to be recognised first. That is why scanned files need the AI mode: it does the recognition step that a plain converter skips, then keeps the field boundaries so a value like 1,488.00 lands in one cell instead of splitting at the comma.

Getting a clean result from a scan

OCR is only as good as the picture it reads, so a sharp, straight scan converts far better than a dim phone photo shot at an angle. If you can, scan or reshoot in good light, keep the page flat, and make sure the numbers are in focus. After exporting, glance down any column of figures for a stray 0 read as O, though clear scans rarely need fixing.

Just need the text, not a table?

If your scan is a page of prose rather than a table, and you only want the words out, our Extract Text from Image extension reads text from any picture or screenshot with on-device OCR. For a PDF table, stick with the steps above.

What about normal, text-based PDFs?

If your PDF is not a scan, you do not need the AI mode at all. Text PDFs convert to Excel or CSV for free. See our guide on converting a PDF table to Excel or converting a PDF to CSV.