PDF file comparison, see what changed.

A PDF file comparison shows what changed between two versions of a document: which words were added, which were removed, and which pages look different. Drop the older file in the Original box and the newer one in the Updated box, and the comparison appears at once. Everything runs in your browser, so contracts and drafts stay on your computer. Use it to check a contract revision, proof a brochure or confirm that an export changed nothing else.

How to compare two PDFs

1. Drop the older PDF on Original PDF and the newer one on Updated PDF. You can also drop both files on either box at once; the first is the original.
2. The cards at the top count the pages that differ, the words added and removed, and the pages added or removed.
3. In Text changes, each changed page shows its words with additions in green and removals in red. Press Next change and Previous change, or the keys N and P, to step through them.
4. The row of numbers is the page pairs. A number such as 2/3 means page 2 of the original is paired with page 3 of the updated file. Click one to jump to it.
5. Switch to Visual to see the pages. Choose Differences, Side by side, Overlay or Slider.
6. Press Download the changes (.txt) for a plain list, or Download a report (PDF) for the changes in colour.

What the visual views show

Differences draws the updated page in pale grey and marks ink that exists only in the updated file in green and ink that exists only in the original in red. A box goes round each changed area. Side by side puts both pages next to each other with the same boxes. Overlay lays the updated page over the original, and the bar under it sets how much of each you see. Slider cuts between them: drag across the page, or use the bar below it.

Visual sensitivity sets how big a change in the pixels has to be to count. Low ignores faint differences, High finds small ones. If the two pages have different sizes, the updated page is fitted into the original and the page says so.

Pairing pages, with an example

By default pages are paired by what is on them. Say the original has two pages, Contract and Schedule, and the updated file has three: Contract, Schedule and a new Annex. The pairs come out as 1 with 1, 2 with 2, and a new page 3. If someone had added the Annex at the front instead, pairing by content would still match Contract with Contract and Schedule with Schedule. Pairing by page number would call every page changed. Switch Pair pages by to Page number if you want strict numbering.

Scanned PDFs have no text to compare

A scan is a picture of a page, so there are no words in it for this tool to read. Pages with almost no text are compared by how they look only, and the page tells you which ones. To compare their words, run each scan through OCR PDF first and then compare the results. Scans also pick up small shifts and dust, which show up as differences, so try Low sensitivity.

What can cause false changes

• Reading order. Words are read in the order the PDF stores them. A page with columns or text boxes can be stored in a different order in two exports, which looks like a change.
• Line breaks and hyphens. The comparison works on words and ignores how lines break. A word split by a hyphen at the end of a line can show as two words.
• Page numbers and dates. A footer that says Page 2 in one file and Page 3 in the other is a real change, so it is shown.
• Fonts and rendering. A new font moves letters by a pixel or two. That shows in the visual views even when the words match.

To combine the changed files afterwards use Merge PDF, and to see the text of one file on its own use PDF to text. For two plain texts, the diff checker works the same way.