Compare PDF

Compare two PDFs and see which pages changed between versions. Text is extracted from both and diffed page by page, with a downloadable report. Runs entirely in your browser — no upload, no sign-up — which matters when the documents are contracts or accounts.

What it compares, precisely

Being exact about the granularity, because it decides whether this tool fits your job.The comparison works at page level. All the text on a page is gathered into one block, and those blocks are compared between the two documents.

So if a single word changed on page 12, page 12 is reported as changed — you are told which pages differ, not which sentence within them. That is less than a word-level diff and considerably more than nothing: on a 100-page agreement, being told that pages 12, 47 and 83 moved turns an impossible review into opening three pages.

The added, removed and unchanged counts follow the same grouping, so read them as a summary of how much of the document shifted rather than an exact tally of edited lines.

How to use it in practice

1. Load the original in the first slot and the revised version in the second. Order matters — it decides what counts as added versus removed.
2. The comparison runs automatically once both files are in. There is no button to press.
3. Read the changed pages from the result, then open just those pages side by side in your PDF reader to find the exact wording.
4. Download the report as a text file if you need a record — minus for removed, plus for added, following the standard diff convention.

What it will not catch

This is a text comparison. Anything that changes how a page looks without changing its words is invisible to it:

The signature row catches people out on contract review: a countersigned copy looks identical to this tool, because the signature is an image rather than text. Check that visually.

Scanned documents need OCR first

A scan is a picture of a page. There is no text in it, so there is nothing to extract and nothing to compare — the tool reports that rather than pretending the pages are identical, which is the failure mode you would actually want to know about.

The fix is to run both files through OCR PDF first, which adds a real text layer while leaving the scanned appearance intact, then compare the OCR'd versions. One caveat worth expecting: OCR is not perfectly deterministic on poor scans, so a handful of reported differences may be recognition variance rather than genuine edits. Treat unexpected single-page differences with mild suspicion and check them by eye.

Why comparing locally matters here

Think about when people compare two PDFs. It is a contract before signing, a policy document after counsel's revisions, a set of accounts against last quarter's, a tender response, a settlement draft. The whole category is material you would not casually upload anywhere.

Both documents are parsed and diffed inside the browser tab, so neither leaves the device. No account, no queue, no retention policy to read, and it keeps working with the connection off once the page has loaded. For this particular tool that is not a nice-to-have — it is the reason to use it over the alternatives.

Related PDF tools

OCR PDF is the prerequisite for comparing scans. To pull out just the pages that changed, extract pages takes a range like 12, 47, 83, andorganize PDF reorders or deletes pages. If you need the text itself rather than a diff, PDF to Word reconstructs editable text with headings and lists, anddiff checker compares two blocks of plain text at line level once you have pasted them in. To mark up the differences you found, edit PDF adds text annotations and add watermark stamps the version status.