How to count words in a PDF (free, private, in your browser)

You'd think counting the words in a PDF would be as simple as selecting all the text and pasting it into a word counter. Sometimes it is. Just as often, though, the “text” isn't really text — it's a picture of text — and your selection comes up empty. Here's the fastest reliable way to get a real word count for any PDF, plus what to do when the simple way doesn't work.

Count words in a PDF in 3 steps

1. Open the PDF Word Counter and drop your PDF onto the upload area. No signup, no upload to any server — the whole thing runs in your browser.
2. Wait a second while the tool extracts the text from every page. Long documents take a couple of seconds; the progress bar shows you where it is.
3. Read the total right below the tool: word count, character count (with and without spaces), and per-page breakdown. Copy the number or export a small summary if you need it.

Why counting PDF words is trickier than counting Word words

A .docx file stores its words as a plain sequence of characters, so counting them is straightforward. A PDF stores text as positioned glyphs on a page — arranged by coordinates, not by reading order. That means:

• Multi-column layouts can read in the wrong order (right column merged into left) unless the extractor respects the visual order. Our counter reconstructs reading order per page.
• Running headers, page numbers, and footnotesget counted alongside body text. If you only want body copy, subtract the header/footer word count once and multiply by page count.
• Hyphenated line-breaks (words split across lines) can look like two words to a naïve counter. Our counter stitches them back together.

What if my PDF is a scan (image)?

Scanned PDFs — for example, ones produced by a phone scanner app or a copier — usually contain images, not text layers. Selecting them in your PDF viewer copies nothing; counting words returns 0.

Fix: run the PDF through SmilingPDF's OCR tool first. It recognises the text inside the images and writes it back into the PDF as a real text layer. After that, PDF Word Counter will read it normally.

Do the numbers match Word's count?

They usually do, within a few percent. Small differences come from how each tool handles hyphenation, contractions, numbers-with-decimals, and things like URLs — different tools apply slightly different word-boundary rules. If you need an exact match with a specific system (e.g. a submission-portal limit), always trust that system's own count.

Is this really free? And is my PDF private?

Yes to both. The counter is part of SmilingPDF's free toolkit — no signup, no ads on the tool itself, no limits on document size beyond what your browser can hold in memory. And because everything runs client-side, the file's text is extracted inside your own browser tab; it never travels to a server.