📊 Word Counter

Why "well-known" and a Whole URL Both Count as One Word

The rule is simple: the counter trims your text, splits it wherever one or more spaces, tabs or line breaks appear, and counts what's left. Nothing without a space inside it gets split, no matter how many separate ideas it actually contains.

don't stop-believing
https://example.com/page 123

4

That match between "one word" and "one whitespace-delimited chunk" holds for almost all ordinary English writing, which is why the number on this page usually feels right without any thought. don't reads as one word because the apostrophe sits inside it rather than around it, and a contraction is exactly what most people mean by "one word" anyway.

It breaks down for text built from pieces that a human would count separately but that never get a space between them. A long product SKU jammed against a price with a slash, a run-on hashtag, or a pasted URL all read as a single word to this counter even though they carry several distinct ideas. This is not a flaw specific to this tool. A plain whitespace split is the same basic approach nearly every lightweight word counter uses, as opposed to a full natural-language tokenizer that understands grammar and dictionary word forms.

Why Chinese, Japanese and Thai Text Barely Registers a Word Count

It is a real blind spot, not a glitch. A whitespace split cannot find word boundaries in a language that does not put spaces between its words, so an entire sentence can register as a single "word." Use the character count for these languages instead.

我们今天去商店买了一些水果和面包

Words: 1
Characters: 16

That sentence means roughly "we went to the store today and bought some fruit and bread," and a native reader would count somewhere around a dozen distinct words in it by that language's own grammar. Since it is typed as one continuous run of sixteen characters with no spaces anywhere, this tool, like most whitespace-based word counters, reports it as a single word. Nothing about the sentence is unusual. This is simply what happens whenever the input language does not use spaces to separate words in the first place.

The character count does not have this problem, since it counts every character regardless of whether spaces separate them, which is exactly why Chinese and Japanese publishing has long used character counts rather than word counts as the standard unit for describing length. A character count sidesteps word segmentation entirely, and word segmentation in these languages is a genuinely harder linguistic problem than splitting on spaces.

Why This Tool, Microsoft Word and Google Docs Never Agree on the Same Paragraph

Different tokenization rules, not a bug in any of them. Each program decides differently where one word ends and the next begins, so the same paragraph can produce three slightly different counts.

None of that makes one number more "correct" than another. There is no single universal standard that all word processors follow, and the small variance rarely matters for everyday purposes: checking a blog post's rough length, hitting a daily writing target, tracking progress on a draft.

Why Ten Emoji Can Report Twenty Characters

"Characters" counts every character you typed, spaces included; "Characters (no spaces)" strips whitespace first. Both numbers can run higher than you would count by eye once emoji are involved.

😀😀😀😀😀😀😀😀😀😀

20

"Characters" counts every single character in your text, letters, numbers, punctuation, spaces, tabs and line breaks alike, treating a blank space exactly the same as a letter. "Characters (no spaces)" strips out every whitespace character first and then counts what remains, so it reflects only the visible, non-space content you actually typed. The gap between the two numbers is a rough proxy for how much of your text is whitespace: a bulleted list full of short lines and line breaks shows a bigger gap than a dense paragraph of continuous prose.

The emoji case above looks like a miscount but is not one. Character counting in JavaScript, which is what runs this page, measures length in UTF-16 code units rather than in the symbols a reader sees. Most letters and symbols take up exactly one code unit, but many common emoji, including 😀, are stored internally as two code units, called a surrogate pair. Ten of them typed in a row report a character count of twenty, not ten, even though a reader sees ten symbols.

Why a One-Word Note Shows the Same Reading Time as a 199-Word One

Both round up to the same full minute. Reading time is your word count divided by 200, rounded up to the next whole minute, and it only ever shows "<1m" when the box is completely empty.

The 200-words-a-minute figure behind that math is a commonly cited average silent-reading speed for an adult reading ordinary prose in their native language, and it is a reasonable default for a general estimate. But the rounding is not gradual. Dividing by 200 and rounding up means any word count from 1 through 200 produces exactly the same result, one full minute, and only a completely empty editor produces anything less than that.

That fixed rate also does not slow down for dense or technical material. A page of legal terms or source code takes most readers noticeably longer to get through than a page of casual narrative at the same word count, and the estimate has nothing to say about images, embedded video or tables mixed into an article, since it only ever counts words. Treat the number as a useful, honest planning estimate, a "3-minute read" label on a blog post, rather than a precise prediction of how long any specific reader will spend on the page.

Why "Dr. Smith Arrived at 3.5pm." Counts as Four Sentences

The sentence counter cannot tell a real sentence ending from a period inside an abbreviation or a decimal. It splits on every run of periods, question marks and exclamation points, full stop.

Dr. Smith arrived at 3.5pm. He was late.

4

A human reader sees two sentences in that line. The counter sees four, because it treats the period after "Dr", the period inside "3.5", and the period after "late" as three separate sentence boundaries rather than recognizing which ones are genuine endings. It has no concept of abbreviations, decimal numbers or initials. It only knows where a period, question mark or exclamation point sits in the text.

That works well for ordinary prose written without many abbreviations, and it is a straightforward, predictable rule rather than a black box. Text loaded with "Dr.", "e.g.", "U.S." or prices like "$3.50" will inflate the sentence count well above what a person would count by hand, in rough proportion to how many stray periods it contains.

Why Three Wrapped Lines Count as Three Paragraphs, Not One

The paragraph counter has nothing to do with blank lines. It counts every line that has content on it, whether or not a blank line separates it from the line before.

This is line one
that continues here
and ends here.

3

It is tempting to assume paragraphs are counted by looking for a blank line between blocks of text, the way a word processor distinguishes a new paragraph from a soft line wrap. That is not what happens here. Three lines meant to read as a single paragraph, each ending in a manual line break instead of wrapping naturally, produce a paragraph count of three, not one, because the counter has no way to know those three lines were meant to stay together.

This matters most for text copied from a source that hard-wraps every line: old email threads, some PDF exports and certain note-taking apps all do this routinely. A "paragraph" in the original document can arrive here as several separate lines rather than one continuous block, and the paragraph count reflects those line breaks rather than the writer's actual paragraph structure.

What Word and Character Count Should You Actually Aim For

It depends on where the text is going. Social platforms cap by character, search fields truncate by character, and blog posts get judged by word count. The numbers for each are below.

These numbers shift over time as platforms adjust their limits, so treat them as a working reference rather than a permanent guarantee, and check a platform's current help documentation before relying on an exact figure for something important.

Common Ways People Use a Word Counter

Hitting an exact assignment or submission length

A student with a "1,500 words, no more, no less" requirement, or a freelancer contracted for a specific article length, needs an exact number rather than a rough sense of "about right." Typing directly into the editor and watching the live count, or setting a target above the editor and watching the progress bar fill in, replaces guesswork with a running total that updates on every keystroke.

Tracking daily writing progress

Writers working through a longer manuscript often set a daily word goal, five hundred words, a thousand words, whatever the project calls for, and paste each day's draft in just to confirm they hit it. The target field on this page is built for exactly that: enter a goal once, and the progress bar and percentage update automatically as the word count changes.

Drafting against a strict character limit

Anyone writing an SEO title tag, a meta description, or a caption for a platform with a hard character cap benefits from watching the character count while drafting rather than writing freely and trimming afterward. Truncation on search results and social platforms usually cuts off mid-sentence with an ellipsis, so knowing the limit before you finish the sentence produces a cleaner result than editing down after the fact.

Estimating translation or localization scope

Translators and localization teams frequently need to estimate cost or effort before starting a project, and both commonly scale with source word or character count depending on the language pair. Running a source document through the counter first gives a concrete number to quote against, rather than an eyeballed guess at document length.

Checking form field and database limits

Developers and content editors populating a field with a known maximum length, a product description column capped at 500 characters, a bio field capped at 160, can paste candidate text in here to confirm it fits before submitting it to a form or an API that might reject or silently truncate an oversized value.

Frequently Asked Questions

Related Text Tools

Need to do more with your text? Try these free tools:

🔒 100% Private & Secure

Your privacy matters. Unlike other word counters that may upload your text to servers, TextSorter processes everything locally in your browser. Your text never leaves your device-we can't see it, store it, or access it in any way. Use our word counter with complete confidence for sensitive documents, confidential work, or personal writing.