Text

Word and character counter

Type or paste your text and watch the counters update themselves. With reading time, keyword density, per-platform limits and a readability score worked out with your language's own formula.

Never sent, never stored

Your text never leaves this page, and reloading leaves nothing behind: it is not stored in the browser either.

Why it matters that the text is not uploaded

What gets pasted into a word counter is rarely public: it is the assignment you have not handed in, the article your client has not signed off, the announcement going out next week, or an internal email you are sizing up before sending. Plenty of counters send every keystroke to their server to do the arithmetic there. Here there is nowhere to send it: the counting happens in your tab. And nothing is stored, not even in your own browser — reload and the box is empty — because with someone else's draft the safest thing is to keep none of it.

How it works

  1. Just paste

    The cursor is already in the text box when the page loads, so you can paste straight in with Ctrl+V without a single click.

  2. The counters run themselves

    Words, characters with and without spaces, sentences, paragraphs and lines all recalculate as you type. No button to press.

  3. Tune what you need

    Pick the limit for whatever platform you are writing for, and check the reading time, keyword and readability panels.

What people use it for

Character count online without the text leaving your machine

Counting words is about the simplest thing a computer does, and yet most online counters send what you type to their server to do the sum there. There is no need: the whole operation fits in the browser, and that is where it stays. Nothing you paste travels anywhere, and nothing is kept between visits — reload the page and the box is empty.

The counters sit above the text box, visible from the first moment and updating as you type. The cursor is already inside when the page loads, so landing and pasting with Ctrl+V is the whole workflow. On a phone the focus is not stolen, so the keyboard does not leap up and cover half the screen uninvited.

How long a text takes to read

Reading time comes in two flavours because they are two different things. Out loud, the limit is how fast people speak: around 150 words per minute in English, 145 in Spanish. That is the number that matters when you are preparing a talk, a video or a podcast against a fixed slot. Silently is much quicker, roughly 240 and 220 words per minute, because nothing has to be pronounced; that is the number that goes at the top of an article.

These are adult averages, not stopwatches. A technical text reads slower than a novel, and someone used to reading aloud runs above the average. They are there to tell you whether your text is three minutes or fifteen, which is usually the question.

Keyword density, minus the words that carry nothing

The keyword panel lists the terms you repeat most, with their count and their share. What matters is what it does not list: articles, prepositions, pronouns and auxiliary verbs are left out, because they turn up in every text and say nothing about the subject. That stop-word list is different for English and Spanish — the same words will not do — and it is picked from the language of the text, not the language of the page.

The percentage is measured against content words, not the whole text. That difference matters: measured against everything, any term lands on a meaningless 1% because it is competing with every “the” and every “of”; measured against content words, the figure can be compared from one piece to another. And accents do not split words: in Spanish, “análisis” and “analisis” are counted together.

The limit for whatever you are writing for

Every platform cuts at a different number, and not remembering them is normal. There are presets for the usual ones — meta description 160, SEO title 60, X 280, LinkedIn 3,000, Instagram caption 2,200 and text message 160 — plus room for your own. The bar changes colour: amber from 90% on, red the moment you go over, with the characters left or the characters over.

And it can count in words instead of characters, which is how academic limits arrive. Social and SEO count characters; universities count words.

Readability: the formula has to match the language

A readability score measures how hard a text is to read from two things: how long the sentences are and how long the words are. For English the standard is Flesch Reading Ease, from 1948. For Spanish there is the Fernández Huerta adaptation, from 1959, which adjusts the coefficients because Spanish words carry more syllables on average.

The difference is not cosmetic: running a Spanish text through the English formula returns a score systematically lower than it deserves, making perfectly readable prose look difficult. It is a very common flaw in tools that only implement Flesch. Here the language of the text is detected, the right formula is applied, and the tool says which one it used. If the detection gets it wrong, you can force it by hand.

Both score 0 to 100, higher being easier. Above 80 is text anybody follows; 60 to 70 is ordinary newspaper prose; below 30 is dense material you have to read twice. One honest caveat: in English syllables are estimated, because no syllable-splitting rule works every time — “create” and “beautiful” both defeat it — so the English score is a guide. In Spanish the splitting is regular and the count is exact.

How each thing is counted

Worth spelling out, because every program does it slightly differently and that is why numbers disagree. A word is a run of letters or digits that may contain a hyphen or an apostrophe: “state-of-the-art” is one word, not four. Characters are counted by visible sign, so an emoji counts as one even though it takes more room underneath. A sentence ends at a full stop, question mark, exclamation mark or ellipsis, without being fooled by decimals or by abbreviations like “Mr.” or “etc.”. A paragraph is a block separated by a blank line; with no blank lines, every line with text counts as one.

If what you want to count is inside a PDF, the PDF toolbox lets you prepare it first; and if you write product copy or metadata from a spreadsheet, the CSV converter can save you the step in between.

And if the text arrives messy — pasted from Word, with double spaces and line breaks in the wrong places — or needs putting into capitals, Title Case or a URL slug, that is what the case converter is for, cleanup panel included.

Frequently asked questions

How exactly is a word counted?

A word is any run of letters or digits, and it may contain a hyphen or an apostrophe. So “state-of-the-art” counts as one and “don't” does too.

Standalone numbers count: “it costs 25 euros” is four words. Emoji and punctuation do not count as words, though they do as characters. Same rule Word uses, so the number should match.

Why doesn't the count match Word exactly?

On words it usually does. The differences are in sentences and paragraphs, because every program draws the line somewhere slightly different.

Here a sentence ends at a full stop, question mark, exclamation mark or ellipsis, ignoring decimals like 3.14 and abbreviations like “Mr.” or “etc.”. A paragraph is a block separated by a blank line; with no blank lines, every line with content counts as one.

Where do the reading times come from?

From average adult rates. Out loud, 150 words per minute in English and 145 in Spanish, because speech sets the pace. Silently, 240 and 220, which is much quicker.

They are averages: a technical text reads slower than a novel, and a fast reader falls outside them. They exist to tell you whether your text is three minutes or fifteen.

What is keyword density and what is it for?

It is the share each word takes of all the content words. It shows what your writing is actually about and catches repetition you had not noticed.

Articles, prepositions and pronouns are left out, with different lists for English and Spanish. The percentage goes against content words rather than the whole text, which is what makes it comparable between pieces.

Which readability formula is used?

It depends on the language of the text, and the tool says which it used. For English, Flesch Reading Ease. For Spanish, Fernández Huerta, the Spanish adaptation of Flesch.

Running Spanish through the English formula returns a systematically lower score than it deserves, because Spanish words carry more syllables. A common flaw in tools that only implement Flesch.

Is the readability score exact?

In Spanish, yes: syllable splitting is regular there and the count is exact.

In English syllables are estimated, because no rule works every time — “create” and “beautiful” both defeat it — so treat it as a guide. Every tool that computes Flesch has this same limitation, though not all of them say so.

Is my text stored or sent anywhere?

Neither. All the analysis happens in your browser and no request carries the text anywhere.

It is not stored in the browser either: reload the page and the box comes back empty. That is deliberate, because what gets pasted here is drafts and internal documents.

Can I set a word limit instead of a character limit?

Yes. Pick “Custom” and switch the unit to words, which is how academic limits arrive. The social and SEO presets are in characters because that is how those platforms count.

How much text can it handle?

Far more than you will paste: 70,000 words — roughly 400 pages — are analysed in full in under a tenth of a second. Above about 200,000 characters the update waits a little longer between keystrokes so typing stays smooth.

Related tools