Someone returns a contract "with a few small changes" and does not say what they were. Comparing forty pages by eye is how clauses get missed. This extracts the text from both files and reports exactly which lines appear in one and not the other.
Your file never leaves your device. It is read and edited by JavaScript running in this tab. Nothing is uploaded or stored.
Enter your details
Result
Fill in the fields above and your result will appear here.
What the PDF Compare does
This is a text comparison, and the distinction matters. It finds wording that differs. It cannot see changes to formatting, fonts, colours, images, signatures or layout — a document reformatted entirely but with identical wording will report as a perfect match.
Lines present in one file but not the other are listed for each side, along with a similarity percentage and a comparison of page counts, line counts and file sizes.
How to use this tool
- Add exactly two PDF files. The first is treated as the original.
- Check the order in the file list and swap them with the arrows if needed.
- Press Compare documents and wait while text is extracted from both.
- Read the similarity figure first, then the lists of lines unique to each file.
How similarity is measured
Text is extracted from both files and split into lines.
Lines present in both are counted as matching:
Similarity = 2 × matching lines
÷ (lines in A + lines in B) × 100
Lines found only in one file are listed separately.
This compares wording, not appearance. Identical text in a
completely different layout scores 100%.
Worked example
Example: two contract versions
| Pages | 14 vs 14 — same |
| Similarity | 98.6% |
| Only in the original | 3 lines — the old payment terms |
| Only in the revision | 4 lines — new terms plus a clause |
A 98.6% match on a contract is not reassuring — it is exactly the case where a small change matters most. The listed lines show what actually moved.
What your result means
100% similarity means the text is identical. It does not mean the documents are identical — formatting, images and signatures could differ entirely.
Below about 95% indicates substantial rewriting rather than a few edits.
Both files reporting no text means they are scans. Text comparison cannot work on images of pages; OCR both first.
Important considerations
- Reordered paragraphs appear as removed and re-added, since the comparison works line by line.
- A change of a single word makes the whole line appear in both "only in" lists — that is normal and shows you exactly where to look.
- Headers, footers and page numbers repeat throughout and can add noise to the comparison.
- For scanned documents, run OCR on both first; otherwise there is no text to compare.
- This does not produce a marked-up document. It tells you which lines differ so you can find them in the originals.
Limitations of this tool
- Text comparison only. Formatting, fonts, colours, images, layout and signatures are not compared.
- Scanned PDFs contain no text and cannot be compared without OCR.
- Line-based comparison, so a moved paragraph shows as both a deletion and an addition.
- No visual side-by-side diff or marked-up output is produced.
Frequently asked questions
Does this show visual differences?
No. It compares the text only. Two documents with identical wording but completely different formatting will report as a perfect match — which is why the distinction is stated clearly throughout.
Why do both lists show the same line?
Because a single word changed within it. Line-based comparison treats a modified line as one removal and one addition, which is what points you at the exact place to check.
Can I compare scanned PDFs?
Not directly — scans contain no text. Run OCR on both files first, then compare the results, bearing in mind that OCR errors will show up as differences.
Which file should I add first?
The original. The second is treated as the revision, which makes the "only in" lists read the way you expect.