How to compare two PDFs: contract diffs and the limits of scans
How TwoThink PDF Compare renders two documents side by side, highlights the text differences, skips unchanged pages, and why scanned files need OCR first.
When a contract changes and you are reviewing the revision, you usually have two versions of the same PDF. An online "compare" site will give you a result, but the originals also leave your computer — and I am not comfortable with that for sensitive clauses.
TwoThink Compare PDF renders both PDFs side by side in the browser and highlights the differences in a text diff panel. Nothing is uploaded.
Which PDFs it works with
| Type | Comparable? | Notes |
|---|---|---|
| PDFs exported from Word / Excel | ✅ | They have a text layer, compared line by line |
| Scans and photographed PDFs | ❌ | No text layer — run them through OCR text extraction first |
| Different page counts | ⚠️ | You are told how many pages one side has extra, and the extra pages are treated as added or removed |
How to use it
- Open Compare PDF
- Drop the old version on the left (PDF A) and the new one on the right (PDF B)
- Turn the pages together and read the text panel below: green = added, red = removed, yellow = changed
- Tick "differences only" to jump straight to the next page that has a change
The canvas above is only a side-by-side preview and does not colour pixels. On complex layouts a pixel comparison is slow and prone to false positives; a text-layer diff is steadier.
How it divides the work with text extraction
- To see the differences between two versions: use Compare PDF
- To get text out of a scan first: use Extract Text from PDF
- To add page numbers before sending it on: use Header, footer and page numbers
All of it happens locally, with no sign-up.