PDF Compare

Older document
Drop PDF or click to choose
Newer document
Drop PDF or click to choose
Document to split
Drop PDF or click to choose
Add documents
Drop PDFs or click to choose — report first, then appendices
Images → PDF
Drop images or click to choose
PDF → images
Drop a PDF or click to choose
Scan → searchable (OCR)
Drop a scanned PDF or click to choose
Document to edit
Drop PDF or click to choose
Document to stamp
Drop PDF or click to choose
Document to compress
Drop PDF or click to choose
– 0 deleted + 0 inserted
100%

Section divides

By default Export produces a single side-by-side PDF. Add divides — manually, or with Auto-detect (appendix/annexure cover pages) — and Export produces one PDF per section instead, each named after its title (your browser may ask to allow multiple downloads).
Compare two revisions of a PDF, side by side.
Drop the older and newer documents above, then hit Compare. Pages are paired side by side with synchronised scrolling — where one document has extra content the other side waits, so matching sections stay level. Hover over text to outline the matching sentence or table cell on the other side; hovering a highlighted change outlines the corresponding changed text instead. Use n/p to jump between changes, h to toggle highlights, and Ctrl+scroll (or +/−) to zoom.
Deleted from older Added in newer
Load a PDF above, then click the page where each section starts — it gets a halo and a name. Ctrl/Shift+click selects pages and ⟳ rotates them; “Aa” on a section page names it from the page's own text.
Add PDFs above — the merged document follows the page order shown here.
Images → PDF Drop images in the left box above — each becomes one PDF page, in the order shown below.
PDF → images Drop a PDF in the right box above to export its pages as images.
Scan → searchable Drop a scanned PDF in the right box above — OCR reads each page and writes an invisible text layer underneath, so the document becomes searchable, selectable and comparable. Pages are never re-rendered: the output looks identical to the input.
Page numbers print small and neat in the margin — pick the corner, the style and where counting starts. Bates numbers print a unique padded identifier on every page (e.g. ABC-000041), the standard for legal and contract document sets. Watermark prints a large label — DRAFT, SUPERSEDED, FOR INFORMATION — across every page. Tick any combination, then Stamp & download. The original file is untouched; stamps are drawn into a copy, and text underneath stays selectable and searchable.
Tool
Load a PDF above, then click any line of text to retype it.
Touch-up edits, not re-typesetting. Click a line of text and retype it — the original is covered and your text is drawn in a matching font, size and colour. PDFs carry no paragraphs, so a longer line never re-wraps the lines below; this tab is for fixing a typo, a date, a name or a number, adding a label, or blanking something out. Add text drops a new text box anywhere (drag to move it); Whiteout covers a region with the surrounding background colour; Image places a picture — a signature, a logo — which you can also paste straight in with Ctrl+V or drag onto the page as a file (drag to move it, corner handle to resize). Edits stay pending while you move between pages — nothing is written until Download edited PDF, and the original file is untouched. Scanned pages carry no selectable text: run Convert → “Scan → searchable (OCR)” on the file first, or use Add text and Whiteout directly. If the document carries a digital signature, any edit invalidates it.
Load a PDF above, pick a method, then Compress.
Smaller images re-compresses the photos and figures embedded in the file — oversized ones are downscaled and everything is re-encoded as quality JPEG — while the pages themselves are never rasterised: text stays selectable, searchable and comparable. This is usually where huge report files shrink the most.

Lossless clean-up rewrites the file with compressed object streams and drops unused objects and stale incremental-save data. Text stays selectable and nothing is re-encoded, so savings are modest unless the file carries dead weight.

The JPEG presets re-render every page as a photo at the chosen detail — this is where huge files (scans, image-heavy reports) shrink dramatically. The trade-off: text in the output is a picture, so it can't be selected or searched, and such pages can't be text-compared later. 200 dpi still prints crisply; 110 dpi is for email-sized copies. If the result wouldn't be smaller than the original, nothing is downloaded.
Select text on the page, then “Use as section name”.
Align