PDF Word Counter

PDF PRIVACY WORD COUNT

About this tool

Word + character + sentence stats

Count words, characters, sentences, paragraphs, and reading time in any PDF. Per-page breakdown CSV.

RT-IMG-021 · Image & File

Drop a PDF here or click to choose Counts words, characters, sentences, reading time

💡 Reading time uses 230 words/min — the standard rate used in publishing for adult readers on dense material. Light content (news, ad copy) reads faster; technical content reads slower.

Drop a PDF to begin.

🔒 PDFs stay on your device. Text extraction + counting happens entirely in your browser via self-hosted pdf.js. Nothing is uploaded.

How to count words in a PDF

Add your PDF

Drag or click to choose. The tool extracts every page's text and computes word, character, sentence, paragraph, and reading-time statistics.

Read the summary

Top panel shows the totals across the document — words, characters with and without spaces, sentences, paragraphs, estimated reading time. Below that is a per-page table for granular analysis.

Export the breakdown

Click Download per-page CSV to save the table as a CSV file. Open in Excel, Sheets, Numbers, or any tool that reads CSV — useful for invoicing per-word translation work, academic word-count compliance, or content audit reports.

Word counting in PDFs — billable, gradable, audit-worthy

Word counting in a PDF is the input to a surprising number of business processes. Translators bill per-word; their first action with any client document is to pull a word count from the source PDF. Academic submissions have strict word-count limits (5,000 word essays, 80,000 word theses); students need defensible counts that match what graders will check. Content auditors costing migration work bill by the page or by the thousand words. Patent filings, legal briefs, regulatory submissions — all involve word counts that matter in money and in compliance.

Why every tool gives a slightly different number

"Word count" sounds simple but actually requires choices. What counts as a word? Hyphenated terms (state-of-the-art) — one word or three? URLs and email addresses — one word or many? Page numbers and headers — counted or excluded? Footnotes? Tables? Different tools answer these questions differently, so Microsoft Word, Adobe Acrobat, Google Docs, and this tool can all give different totals for the same PDF. The right answer depends on your contract or your style guide — and the right defence is to be transparent about the rule applied.

For billable word counts, what matters isn't the tool — it's the rule. Pick a tool, document it in your engagement letter, stick with it. The variance between tools is 1-3% on most documents.

The APAC use case

Word counting is heavy across Singapore and Hong Kong's translation industries (per-word billing for legal and financial document localisation), Vietnam and the Philippines' content-writing and BPO sectors (deliverable accountability), Indonesia's thesis-checking ecosystem (university submission compliance), Malaysia's academic and government workflow (grant proposal word limits), and across Thailand's tourism marketing (content audit for SEO compliance).

What this tool counts — and how

Words are whitespace-separated runs of non-whitespace characters after collapsing repeated spaces. Hyphenated compounds count as ONE word. URLs count as one word. Characters are the raw character count, with a separate "no spaces" total for the standard publishing metric. Sentences are runs ending in ., !, or ? followed by whitespace — a deliberate naive count that won't fool itself on "Dr. Lee" but also won't be perfect either. Paragraphs are separated by blank lines. Reading time uses the publishing standard of 230 words per minute.

10 Things to Know About Word Counts

The average English word is 4.7 characters. So "characters / 5" is a quick mental approximation of word count when you only have a character total.

Microsoft Word, Adobe Acrobat, and Google Docs all use slightly different word-counting rules. Differences of 1-3% on the same document are normal.

Professional translators charge per word in the SOURCE language. A French → English translation is billed against the French word count, not the English output.

Chinese, Japanese, and Korean have no word delimiters. Word counts for those languages are estimated by character count divided by an average characters-per-word (roughly 1.7 for Chinese, 2.5 for Japanese, 2.2 for Korean).

The "230 words per minute" reading-time standard is the median for adults reading prose. Technical material is closer to 150 wpm; light news is closer to 280 wpm.

A novel is roughly 80,000-100,000 words. A PhD thesis ranges from 50,000 (sciences) to 100,000+ (humanities). A blog post sweet spot is 1,500-2,500 words for SEO.

Academic word limits typically include footnotes but exclude bibliography. Different journals have different rules — check the style guide before submitting.

Tables in PDFs usually contribute meaningless "words" (column headers, cell values) to the total. For pure-prose word counts, exclude pages with tabular content manually.

Scanned PDFs have ZERO word counts — there's no extractable text. If the count returns 0 on a document that visually has text, run OCR first.

Per-page word distribution is a useful editing signal. Pages with 80%+ of average density are typical prose; pages with <30% are usually section breaks, illustrations, or tables.

FAQ

No. pdf.js runs entirely in your browser; counting is pure JavaScript. Your PDF is read into memory, analysed in memory, results shown in the panel. Open DevTools → Network and watch — zero outbound traffic.
Every word counter uses slightly different rules — what counts as a word, whether to include hyphenated terms as one or many, how to handle URLs and numbers. Differences of 1-3% are normal across tools. For billable purposes, pick one tool, document it, stick with it.
The PDF is a scanned image with no embedded text. There's nothing to count. Run the file through OCR first — Adobe Acrobat's "Recognize Text" tool or macOS Preview's "Searchable PDF" export — then re-load it here.
Yes. The tool extracts ALL text from each page, including headers, footers, and page numbers, and counts everything. For a "body text only" count, run the document through your text-extractor (the other RECATOOLS tool) with headers off, then count manually in a text editor.
The character counts are accurate for CJK scripts. The "word count" will be misleadingly low because CJK languages don't use spaces between words. Use the CHARACTER count for those languages — it's the standard billing unit for CJK translation work.
It uses the naive rule: count runs of ., !, or ? followed by whitespace. Works well for most prose. Will mis-count when sentences end in abbreviations ("Dr. Lee said.") or contain mid-sentence dot-separated tokens (URLs, version numbers). Treat it as a rough indicator, not a contract metric.
230 words per minute, rounded up to the next minute. Standard publishing-industry rate for adult readers on dense prose. Light material reads faster (280 wpm); technical material reads slower (150 wpm). Adjust your expectation accordingly for your content.
Yes — the Download per-page CSV button saves a CSV with columns: Page, Words, Characters, Characters (no spaces), Sentences, Paragraphs, Reading time. Plus a TOTAL row at the bottom. Open in Excel, Sheets, or Numbers.
Soft limit: browser memory. Desktop browsers comfortably handle 500+ MB / 1000+ pages. The counting itself is fast — bottleneck is pdf.js text extraction (~30 pages/second on a typical laptop).
Yes on iOS Safari and Chrome on Android. 100-page documents typically count in 3-8 seconds on a phone. The stat grid reflows to 2 columns on phone-width screens.

PDF Word Counter