Count the words in a PDF, page by page
Translators quote by the word, publishers set limits by the word and students submit essays against a word count — but a PDF has no word counter built in. This PDF word count tool counts words, characters with and without spaces, sentences, paragraphs, pages and reading time, for the whole file and for every page.
A raw copy-and-paste from a PDF inflates the count: every running header, footer and page number is repeated on every page, and words split by a hyphen at the end of a line become two. The counter rebuilds the text first — lines and paragraphs from the position of the text on the page, hyphenated words joined, sentences that run over a page break kept together — and leaves out repeated headers, footers and page numbers unless you want them counted.
Drop several PDFs at once for a batch table with totals and a CSV export. Scanned pages can be read with OCR, and everything stays on your device.
How to count words in a PDF
- Drop one or more PDFs onto the upload area, or click it to choose files. Counting starts straight away.
- Read the total in the large result, with characters, sentences, paragraphs, pages and reading time below it.
- With several files, use Show counts for to switch between the total and a single file; the per-page table appears for a single file.
- Untick Leave out repeated headers, footers and page numbers if you need every word on the page counted.
- For scanned pages, tick Read scanned pages with OCR and choose the language. Then download the results with Download CSV or Per-page CSV.
PDF Word Count features
Same rules as the Word Counter
Words, characters, sentences and paragraphs are counted exactly as in the Word Counter, so numbers match when you compare.
Headers and footers excluded
Lines in the top or bottom 15% of the page that repeat on at least 40% of the pages, and bare page numbers there, are left out and listed so you can see what was removed.
Per-page breakdown
Words, characters and sentences for every page, with a note on pages that have no text layer.
Batch mode with CSV
Count many PDFs in one go, see a table with totals and export it, or export every page of every file.
OCR for scans
Pages without a text layer can be read with Tesseract OCR in 16 languages, including Hindi, Bengali, Tamil and Arabic.
Check what was counted
Open “Show the counted text” to see the exact text behind the numbers.
How words, characters and sentences are counted
A word is any run of letters or digits in any script; hyphenated words such as “well-known” and contractions such as “don’t” count once, and numbers like 2025 count as words. Characters include letters, digits, punctuation, spaces and the line breaks between paragraphs; characters without spaces leave out all whitespace. A sentence ends with a full stop, question mark, exclamation mark or ellipsis, and a paragraph is a block separated by a larger gap between lines.
Reading time assumes 238 words per minute, the average silent reading speed for adults, and speaking time 150 words per minute, a comfortable presentation pace. The totals for a file are counted on its whole text, so a sentence that continues onto the next page counts once; per-page figures count only what is on that page.
| What is in the PDF | How it is counted |
|---|---|
| Running header “Annual Report 2025” on every page | Left out (option on) and listed |
| “Page 3 of 12” in the footer | Left out as a page number |
| “infor-” at a line end + “mation” on the next line | Joined: 1 word |
| Table cells | Counted like any other text |
| Text inside images (scans, screenshots) | Only with OCR turned on |
Word counts for translation quotes and submissions
Translation agencies in the UK, US and India commonly price per source word, often with a minimum fee, so a reliable count of the source document matters. The CSV export lists every file with pages, words and characters, ready to paste into a quote or invoice. For languages written without spaces, such as Chinese and Japanese, quote by characters instead of words.
For an essay or dissertation, check what your institution counts. Many exclude the reference list, appendices and footnotes — open the per-page table to subtract those pages. To count a passage, copy it from PDF to Text into the Word Counter.
When the count looks wrong
- Zero words on a page means it has no text layer — usually a scan. Turn on OCR, or make the file searchable first with PDF OCR.
- Words run together or split oddly when a PDF positions single letters individually or uses fonts without a proper character map. OCR on that page is more reliable.
- Fewer words than expected? A line you wanted might have been treated as a header or footer — untick the option and compare.
- Need the number of pages only? PDF Page Count counts pages in many files at once.
PDF Word Count FAQ
How do I count the words in a PDF?
Open the PDF on this page. The total number of words appears at once, with characters, sentences, paragraphs, pages and reading time. It is free, needs no sign-up and has no file limit.
Is the PDF word count the same as Microsoft Word?
Very close. Words are counted with the same rules as Word and our Word Counter: hyphenated words and contractions count once. Differences come from what is included — Word counts its own headers, footnotes and text boxes differently, and this tool leaves out repeated headers and footers by default.
Can I count words in a scanned PDF?
Yes. Tick “Read scanned pages with OCR” and choose the language. Pages without a text layer are then read with OCR; the count is an estimate because OCR can misread a few words.
Can I count words in several PDFs at once?
Yes. Drop all the files together. You get a table with pages, words and characters per file, a total, and a CSV export for your records or a translation quote.
Does it count headers, footers and page numbers?
Not by default. Repeated lines at the top or bottom of the pages and bare page numbers are left out and listed. Untick the option to include them.
Is my PDF uploaded?
No. The text is read and counted in your browser. With OCR on, only the OCR engine and its language data are downloaded; your file never leaves your device.