text tool
Word counter
This one is a writing surface rather than a box with a number under it. The counts sit along the foot of the page and update on every keystroke, and the text itself picks up marks where a sentence runs long, where a verb has gone passive and where a long word has a shorter twin. It all happens in this page, so a draft contract or somebody's application never leaves your device.
Runs on your device. Nothing is uploaded.
Loading the tool…
How to use it
- 1
Write or paste into the page
The whole panel is the writing surface. Counts appear along its foot straight away and follow every edit.
- 2
Read the marks in your own text
A solid or double underline is a sentence that makes the reader work. Dotted is an adverb, dashed is passive voice, wavy is a long word with a shorter everyday twin.
- 3
Check the reading score
The big number is Flesch reading ease, from 0 to 100. Sixty and above is plain English. The band underneath shows where your text sits.
- 4
Open the frequency panel if you repeat yourself
It ranks the words and phrases you use most, with common filler left out, plus a paragraph by paragraph breakdown and a check against common length limits.
What the highlights actually mean
Each sentence gets its own Flesch-Kincaid grade. At grade ten or above it is marked as hard, at fourteen or above as very hard, and only sentences of ten words or more are eligible so a short punchy line is never flagged. Adverbs are words ending in ly, with a list of exceptions such as reply and family. Passive voice is a form of to be followed by a past participle. Long words are matched against a list of about eighty terms that have a plainer equivalent, so utilise becomes use and commence becomes start. Every mark carries a different underline as well as a different colour, so the page still works if you cannot tell the colours apart.
Reading time and speaking time are not the same
Silent reading runs at roughly two hundred words a minute for ordinary prose. Reading aloud at a pace an audience can follow is closer to a hundred and thirty, and slower again if you want pauses to land. That is why a three minute talk is about four hundred words rather than six hundred, and why scripts written to a reading estimate always overrun. Both paces are adjustable in the advanced options if you know your own.
Why characters without spaces is a separate number
Because most limits count them differently. Print and design work is usually specified in characters including spaces, since that is what fills a line. Some legacy systems and a few academic limits count without. When a limit matters, check which one is meant before you start cutting, because on a long paragraph the two numbers differ by roughly a sixth.
Nothing is uploaded, and you can check
The counting and the highlighting happen in JavaScript in this page. After the page loads it makes no network request, which you can confirm in your browser's network tab: type into the box and watch nothing happen. That is worth knowing for the common case where the text you are counting is confidential and pasting it into a random website would be a problem.
Limits
- Sentence counting uses full stops, question marks and exclamation marks, with a short list of abbreviations such as Dr. and e.g. skipped, so unusual abbreviations can still split a sentence early.
- Syllables are counted with a vowel group heuristic, not a dictionary, so the Flesch numbers are accurate to roughly a point rather than exact.
- Passive voice and adverbs are found by pattern, not by grammar, so a few will be wrong in both directions.
- Above 120,000 characters the highlighting and the reading score pause so typing stays smooth. The counts stay live.
- Reading time assumes 200 words a minute and speaking time 130, which are averages rather than your pace.
What it does not do
- It does not check spelling or grammar.
- It does not rewrite anything for you, it only shows you where to look.
- It does not save your text unless you tick the box in the advanced options.
- It does not read an uploaded file; paste the text instead.
Questions people ask
Is my text uploaded anywhere?
No. Everything runs in your browser and the page makes no network request after it loads. You can verify that in the network tab of your browser's developer tools: type into the surface and nothing is sent. Refreshing clears the text, because nothing is stored either unless you tick the box in the advanced options.
How is the reading ease score calculated?
With the standard Flesch formula: 206.835 minus 1.015 times the average sentence length, minus 84.6 times the average syllables per word. The grade figure uses Flesch-Kincaid, which is the same two inputs weighted differently. Both were built for English prose, so they mean very little for code, lists or a language other than English.
How are syllables counted?
By a common heuristic rather than a dictionary. Words of three letters or fewer count as one. Otherwise a silent trailing e is removed, a leading y is dropped, and the remaining vowel groups are counted, with a floor of one. It is right most of the time and wrong on names and unusual spellings, which is why the score should be read as a guide.
Why is a sentence marked hard when it reads fine to me?
Because the mark is about effort, not correctness. A sentence is flagged when its own grade level reaches ten, which usually means it is both long and full of multi syllable words. Plenty of flagged sentences are perfectly good. Treat the marks as places worth a second look rather than errors to clear.
Why does the sentence count look wrong?
Sentences end at a full stop, question mark or exclamation mark followed by a space or the end of the text. A short list of abbreviations such as Dr. and Inc. is skipped, and so are single initials, but an abbreviation outside that list will still split a sentence early and push the count up by one.
Can I count words in a PDF or a Word file?
Not directly. Open the file, select the text and paste it here. That keeps the tool simple and keeps your file on your own machine, which is the trade this whole set of tools is built around. If you need the text out of a PDF first, the PDF text extractor on this site does that in the browser too.
Does it count words the same way as a word processor?
Very nearly. Words are runs of letters and digits, with apostrophes and hyphens kept inside a word, which is what most processors do. Chinese, Japanese and Korean characters are counted one per character, as those languages have no spaces. Expect a count within a word or two of Word rather than an exact match.
Turn this text into a finished document
Take what you made here into documents.
Create a document with OneCraftRelated tools
- Case converterSee sentence case, title case, camelCase, snake_case and eleven more rendered from your own text, then press the one you want to copy it. Runs in your browser.
- Text diffCompare two versions line by line and word by word, with side by side or one column view. Runs in your browser, so a draft contract never leaves your machine.
- Speaking time calculatorPaste a script and see it laid out on a clock, section by section, against the slot you have. Six measured speaking rates, and a rehearsal timer.
- Social character counterWrite the post once and see it as the feed will show it, with the limit on the preview, the fold marked, and a thread split when it runs long.
- Lorem ipsum generatorPlaceholder text drawn as a page, a card grid, headlines, a list or rows of mock data. Latin or plain English, varied lengths, copies as HTML or Markdown.
Background reading: the question page on this.
Written and checked by the OneCraft team. Last checked .