The tool extracts the text of each page — grouped under --- Page N --- markers — and diffs the two extracts. That makes it ideal for revisions of contracts, policies and reports where wording is what changed. It will not flag a re-laid-out page, a changed font colour or a swapped logo, because those are not text changes. Scanned documents have no text layer at all: for those, the tool says so plainly — run OCR elsewhere first if you need a comparison. The output is capped at 500,000 characters for very large files.
Checking what actually changed between two exports of the same report, verifying that a redacted draft really dropped the sensitive clauses, or confirming a vendor sent a revised rather than identical agreement. Both extractions and the diff run entirely client-side, so confidential drafts stay on your machine. When you then need to act on the result — removing the changed pages, or re-ordering the final set — the rest of the PDF suite picks up where this leaves off. The diff itself is plain text in a scrollable panel, so copying a hunk into a review comment or a ticket is trivial.
Also useful: PDF to Text Extractor, Text Diff, and Organize PDF.