Word Counter
Count words, characters, sentences and paragraphs online for free.
Advertisement
64
Words
379
Characters
315
No Spaces
5
Sentences
2
Paragraphs
3
Lines
Read Time
19s
~12.8 words/sentence
Readability
Difficult
Flesch-Kincaid score: 50 / 100
Longest word: FreeToolKitio
Advertisement
How to Use This Tool
Paste or Type Your Text
Click the text area and paste your document, essay, or any text content. The counter updates in real-time as you type.
View Your Stats
See instant counts for words, characters (with and without spaces), sentences, paragraphs, and estimated reading time.
Clear and Count Again
Click the Clear button to reset and start counting a new piece of text. There are no limits on how many times you can use it.
Advertisement
Related Tools
Frequently Asked Questions
Does the word counter count words accurately?
How is reading time calculated?
Can I count words in different languages?
Is there a character limit for counting?
About Word Counter
A freelance journalist is finalizing an op-ed for the Guardian's 'Comment is Free' section where the editor quoted a firm 900-word ceiling, and Google Docs insists the draft is at 903 while the submission form is going to reject anything over budget. A graduate student is trimming a dissertation chapter to stay under a supervisor's 8,000-word limit that excludes footnotes and the bibliography. This counter runs against the live document in your browser, updating on every keystroke so you never have to hit a 'recount' button. It reports words (split on Unicode whitespace and punctuation, matching the algorithm Microsoft Word uses), characters with and without spaces, sentences (split on terminal punctuation followed by whitespace), paragraphs, and reading time at 238 words per minute — the median adult silent reading speed per a 2019 meta-analysis by Brysbaert. There is also a speaking time estimate at 150 wpm for presentation scripts and a keyword density table so you can see if you accidentally used 'leverage' eight times in a 400-word paragraph.
When to use this tool
Hitting a hard editorial word budget
An essayist submitting to a publication with a strict 1,200-word limit needs live feedback as cuts are made. Watch the counter drop as sentences get tightened, and the reading-time estimate stays honest for the editor's intro blurb (roughly 5 minutes at target length).
Staying under Twitter or LinkedIn platform caps
Twitter allows 280 characters, LinkedIn posts cap at 3,000 characters, and a LinkedIn headline at 220. Paste a draft, watch the character count, and trim from the least load-bearing clause when the red-line fires rather than discovering the truncation after posting.
Meeting a university submission requirement
A thesis chapter must be 8,000 to 10,000 words excluding footnotes, references, and appendices. Paste the body text alone to confirm the number lands inside the window before uploading to Turnitin, which will reject over-length submissions without warning.
Estimating a conference talk duration
A 20-minute conference slot roughly fits 2,800 to 3,000 spoken words at a comfortable pace. Paste the draft script, read the 150-wpm speaking time estimate, and trim anecdotes or cut a section if the total blows past the slot allotment before dress rehearsal.
Auditing keyword density for SEO copy
A long-form blog post targeting 'freelance tax deduction' should mention the phrase 3 to 6 times across 1,500 words without tripping into keyword stuffing. The density table flags when a single term exceeds 2 percent of the word count — the threshold Moz and Ahrefs flag as spammy.
How it works
- 1
Words split on Unicode whitespace and punctuation
We split the input string with a regex that matches any Unicode whitespace (\s) or word-boundary punctuation, then filter out empty tokens. This matches Microsoft Word and Google Docs within 1 to 2 words on typical prose. Hyphenated compounds like 'state-of-the-art' count as one word; em-dash-separated clauses count as two.
- 2
Sentences detected by terminal punctuation
A sentence break is a period, exclamation point, or question mark followed by whitespace and a capital letter (or EOF). Abbreviations like 'Dr.' and 'Mr.' are handled with a short exception list, but obscure ones (St. Louis, e.g.) can occasionally inflate the count by one. Sentence count is always within 5 percent of a human tally for normal prose.
- 3
Reading time anchored to empirical research
238 wpm is the median silent reading speed for adults on general prose, from Brysbaert's 2019 meta-analysis of 190 studies. Technical prose (law, science, dense reference material) reads slower at around 150 wpm; light fiction runs faster at 260. We stick with 238 as the general-purpose default and round to the nearest minute.
Pro tips
Paste plain text to avoid hidden Word artifacts
When copying from Microsoft Word or Google Docs, the HTML clipboard payload includes tracked-change markers, comments, and sometimes field codes that inflate character counts. Paste into a plain-text editor first (Ctrl+Shift+V in most editors, or a tool like text case converter) to strip formatting, then paste here. You'll see a character count that matches what the submission form actually receives, not a padded HTML version that quietly fails validation at upload.
Use character count (no spaces) for Twitter threads
Twitter counts URLs as 23 characters regardless of actual length and emojis as 2 characters each. The character count (no spaces) value here is not a direct match, but it is the closest proxy for platform caps on tight-budget posts. For a serialized thread, split at natural sentence boundaries around 240 characters each, which leaves buffer for the '(1/n)' markers and the inevitable last-minute rephrasing when you discover a grammar issue three tweets in.
Treat reading time as a headline asset, not filler
Medium, Substack, and most modern publications display reading time prominently near the title because it is the strongest predictor of whether a reader starts the article at all. Articles labeled '7 min read' land in the sweet spot for in-depth pieces (1,500 to 1,800 words), while '12 min read' drops click-through rates by roughly 30 percent per internal testing shared by publications like Nieman Lab. Use the estimate to target 4 to 8 minutes for your primary audience.
Frequently asked questions
Why does your word count differ slightly from Microsoft Word or Google Docs?
Different word processors use slightly different tokenization rules. Microsoft Word counts hyphenated compounds (state-of-the-art) as one word, while LibreOffice splits them into four. Google Docs treats em-dashes as word separators, while Pages does not. Our counter uses a Unicode-aware whitespace-and-punctuation split that aligns most closely with Microsoft Word. For a 1,000-word document you should see at most a 10-word difference, well under the tolerance any editor will care about. If your submission system uses Word's count specifically, open the draft in Word once before submitting to confirm.
How is reading time calculated and can I trust it for planning a presentation?
Reading time is computed at 238 words per minute, the median silent reading speed for adults on general prose from Brysbaert's 2019 meta-analysis of 190 studies. For silent reading of blog posts or articles, the estimate is accurate within 20 percent for most readers. For presentations or voiceover scripts, switch mentally to the speaking-time estimate at 150 wpm, which matches TED Talk pacing. Dense technical material (legal briefs, medical studies) reads slower at around 150 to 180 wpm silent; light fiction runs faster at 260 wpm. Use the estimate as a ballpark, then time yourself reading aloud for anything time-critical.
Does this counter work on non-English text, and how accurate is it for CJK languages?
For Latin, Cyrillic, Greek, and Arabic scripts, the tokenizer works correctly because those scripts use whitespace between words. For Chinese, Japanese, and Korean (CJK) text, the concept of 'words' is ambiguous because the scripts do not use spaces between characters. We fall back to counting characters for CJK, which matches the convention used by most Chinese and Japanese publishers (who measure length in characters rather than words). A 1,000-character Japanese essay is roughly equivalent to a 500-word English essay in reading time and information density.
Can I get keyword density for SEO optimization without a paid tool?
Yes, the keyword density panel shows the top 20 most-used words along with their frequency and percentage of total word count. For SEO, target keyword density is typically 1 to 3 percent for primary keywords — meaning your 'freelance tax deduction' phrase should appear 10 to 30 times in a 1,000-word post. Anything above 4 percent reads as keyword stuffing and will trigger algorithmic penalties on Google. The panel also helps you spot unintentional overuse of filler words like 'basically' or 'leverage' that weaken prose.
Are my documents saved or analyzed on your servers?
No. All counting happens in JavaScript inside your browser tab. There is no fetch, no form submission, and no logging of content — the text you paste never leaves the page. This matters if you are working with unpublished manuscripts, confidential client drafts, pre-embargo press releases, or legal work. The code is visible in your browser's developer tools under Network (you will see zero requests during counting) and Sources (the tokenizer is readable JavaScript). For sensitive work, you can also disconnect from Wi-Fi before pasting and the tool will continue working exactly the same.
Honest limitations
- · Word-count algorithms differ subtly between Word, Google Docs, and LaTeX, so expect 1 to 3 percent variance from whatever your editor reports.
- · Sentence detection stumbles on texts with heavy abbreviation use (legal briefs, scientific papers with 'e.g.' and 'i.e.' scattered through), occasionally overcounting by a few percent.
- · Reading-time estimate assumes average adult silent reading of general prose; adjust expectations for technical, poetic, or translated content where cognitive load is higher.
Word counting sits next to most writing and editing tools in this set. The text-case-converter is the fastest way to normalize mixed-case copy before a final count, and the duplicate-line-remover cleans out accidentally-pasted repeats that inflate counts in outline drafts. For tightening prose before checking length, the grammar-checker catches redundant clauses you can cut to fit a word budget, and the paraphrasing-tool rewrites verbose sentences into tighter versions when you need to lose 50 words without losing meaning. Long-form authors often pair counting with the ai-writing-assistant for expansion or compression in-place.
Advertisement