PDF compare example: 30 days versus 45 days
A one-clause contract edit, what text extraction can show, and why that is not a visual proof.
Open PDF Compare to run this case yourself.
Version 1, extracted sentence
Payment shall be made within 30 days.Version 2, extracted sentence
Payment shall be made within 45 days.The problem
Two PDF revisions of a short commercial clause look the same on the page. The only wording change is the payment window: 30 days becomes 45 days. That is a commercial change. A logo move or a stamp on the same page would be a different kind of change, and this tool is not built to see that second kind.
What PDF Compare can detect
The tool parses each PDF and reads the text layer. It then collapses extra spaces, rejoins line wraps inside a paragraph, and splits on sentence-ending punctuation (. ! ? ;). These two sentences become two lines. The line diff shows the first as removed and the second as added, with 30 and 45 highlighted inside the pair.
If the same sentence is the only text in both files, that is the whole diff. Headers, footers, and page numbers that are part of the text layer can still appear as extra lines. The comparison is of extracted words, not of the page image.
What it does not detect
Text comparison is not visual or pixel comparison. A changed signature image, a highlighted box, a moved table, or a font-only change will not appear unless the extracted words change. Scanned pages with no text layer are not OCR’d. If you cannot select the sentence in a PDF reader, PDF Compare will not invent it. Encrypted or damaged files fail extraction and show a short message; the parser’s raw error stays in the application log.
When this tool may not be the right choice
- Scanned PDFs without a text layer, or handwriting.
- Visual or pixel comparison, including image-only differences.
- You already copied the paragraphs out. Text Compare skips extraction and compares the paste directly.
Processing and privacy
PDF, Word, and Excel comparison run on the server for the request that submits the form. Each file is read from PHP’s temporary upload path. CompareStack does not copy the upload into a user account, a document database, or application storage. PHP deletes that temporary file at the end of the request when the application has not moved it. The result is the HTML response; there is no saved comparison history. Application logs can still contain the IP address, the requested URL, and—if extraction fails—the exception message. That is not the same as storing the file.
The Privacy Policy and how processing works are the full description.