PDF Compare: Find the Differences Between Two PDF Documents Online

A lawyer gets back a redlined contract and needs to know exactly what changed from the version they sent out. An editor gets a "final" draft and wants to confirm it actually matches the version that was approved. A student compares two versions of a syllabus PDF to see what the professor updated. In every case, the real question is the same: what's actually different between these two files, and scrolling through both side by side hoping to spot it isn't a reliable method.

This is also exactly the kind of task people reflexively upload to a random "compare PDF" website, without pausing to consider that contracts, academic drafts, and internal documents are often the last thing you want sitting on an unknown server, even temporarily.

Why document comparison shouldn't involve a server at all

Comparing two PDFs is fundamentally a text-diffing problem: extract the text from each document and compare it, line by line or word by word, to highlight additions, deletions, and changes. None of that requires a remote server; it's exactly the kind of computation a browser can do using the PDF's own text layer. Keeping the comparison client-side matters most for documents that are confidential by nature, legal agreements, HR documents, unpublished manuscripts, since even a "we delete your files after an hour" promise from a server-based tool is still a promise you have to trust rather than a guarantee.

How to compare two PDFs with ihatetools

Open the PDF Compare tool.
Upload (select, not transmit anywhere) your original document and the revised version.
The tool extracts text from both PDFs and runs a diff, highlighting additions, deletions, and changed sections.
Scroll through the highlighted differences rather than the raw documents; this is usually dramatically faster than manual side-by-side reading, especially for long contracts.
If you need to document the specific changes for someone else, note the highlighted sections or take screenshots of the relevant diffs.

Because the text extraction and diffing both happen locally, you can compare documents immediately after receiving them, without any wait for server processing.

What PDF comparison can and can't tell you

Text-based diffing catches: added or removed sentences, reworded clauses, changed numbers or dates, reordered paragraphs that shift surrounding text.
It won't reliably catch: purely visual changes, like a different font, color, or layout, if the underlying text content is identical. If you need to confirm visual fidelity rather than textual content, you'll need to review the rendered pages directly.
Scanned PDFs are a special case: if either document is a scanned image rather than real text, there's no text layer to diff. Run it through OCR PDF first to create a searchable text layer before comparing.

Practical tips for reviewing document changes

Compare the exact two versions you mean to, it's easy to accidentally diff an old draft against another old draft when multiple versions are floating around in an email thread. Check file names and modification dates first.
For contracts and legal documents, pay close attention to numerical changes (dates, amounts, percentages); these are easy to skim past in a wall of highlighted text but are often the most consequential edits.
If a document has gone through many revision rounds, compare the original against the final version rather than trying to track every intermediate draft; the cumulative diff tells you what actually changed overall.
Once you've confirmed the final version is correct, consider adding a watermark marking it as final with Add Watermark, or locking it down with Protect PDF before distributing.

FAQ

Can PDF Compare detect changes in scanned documents?

Only if the scanned document has already been through OCR to create a text layer. A pure image-based PDF has no extractable text to diff. Use OCR PDF first to convert scanned pages into searchable text.

Will formatting changes show up as differences?

Text-based comparison focuses on content, not visual styling. A change in font or color alone, with identical text, typically won't register as a difference, since the underlying text content hasn't changed.

Is it safe to compare confidential contracts with this tool?

Yes. Both documents are processed entirely in your browser; nothing is uploaded to a server at any point during the comparison.

How is this different from just reading both PDFs side by side?

Manual side-by-side reading is error-prone for long documents, small wording changes are easy to miss. An automated text diff highlights every addition, deletion, and modification explicitly, so you're reviewing flagged changes instead of scanning entire pages for differences.

Spot the changes instantly

Don't rely on eyeballing two long documents side by side. Open PDF Compare on ihatetools and let an automated text diff show you exactly what changed, without ever uploading either file.