Free Text Tools: Word Counts, Case and Cleanup

Text

Text tools answer the questions you have about a piece of writing before you send it anywhere: how long it is, whether it fits the limit you were given, and whether the lines you pasted are the lines you meant to paste.

Word and character counts

What a word and a character actually are

Every count depends on a rule, and most disagreements about word counts are really disagreements about the rule. A program that splits on spaces reports a whole sentence of Chinese as one word, because Chinese is not written with spaces. A program that splits on punctuation counts "don't" as two words, because the apostrophe looks like a separator. Neither is lying; they are answering slightly different questions.

The word counter on this site writes its rules on the page: Chinese, Japanese and Korean count by character, other scripts count by runs of letters and digits, apostrophes and hyphens join a word instead of splitting it, and characters are Unicode code points, so an emoji is one character rather than two. That is why the figure here can differ from your editor by a few words — and why you can work out which of the two fits the limit you were actually given.

The limits that catch people out

Most published limits are stated in the unit that is easiest to count and hardest to agree on. A form asking for a set number of words is counted by a reader using a different tool. A field that says "characters" usually means code points, not bytes, so a sentence in Chinese and the same sentence in English do not look alike to it at all. A resume has no limit written down anywhere and still gets skimmed in seconds.

What a tool can do here is make the number visible so you can decide. The word and character counter reports words, characters with and without spaces, sentences, paragraphs and lines at the same time, plus a reading time you set to your own speed — because "how long will this take to read" has no single correct answer, and a page that prints one as a fact is inventing it. What none of these tools do is judge the writing: length is a measurement, not a verdict.

What belongs in this category next

The rest of this category is string handling with nothing behind it: change the case of a heading, strip the duplicate lines out of a list you pasted from a spreadsheet, sort what is left, compare two versions of a paragraph to see what actually changed, and turn tab-separated text into a Markdown table.

None of those need a data table, a language model or a server, which is the test everything on this site has to pass. What they do need is precision about what they claim: a case converter has to say whether it is changing the first letter of each word or of each sentence, and a comparison tool has to say whether it is comparing lines or characters. Getting that sentence right is most of the work.

Which one do you need?

• Is this too long for the word limit I was given? Use the word and character counter and read the word figure next to the counting rules it states, so you know which rule produced it.
• How long is this in characters, with and without spaces? The same tool prints both, along with sentences, paragraphs and lines, so you can match whichever unit the form is asking for.
• How long will this take someone to read? The counter estimates reading time and lets you set the speed yourself, rather than assuming one number for every reader.
• Is my resume too long, or missing the right words? That is a different question with a written answer: the ATS-friendly resume template covers length and layout, and the free ATS keyword check compares your resume against a posting.
• What percentage of the limit am I at? Use the percentage calculator in the calculators category: your current length as a percentage of the maximum.

Other tools and guides on this site

Text tools cover the shape of what you wrote; these pages cover what goes into it.

• ATS-friendly resume template — a single-column skeleton and the six rules it follows.
• Resume summary examples — for when the opening two lines are the weak part.
• Cover letter template — the four paragraphs worth writing, and the four to cut.
• All tools — every calculator and utility on the site, grouped by what it is for.
• Calculators — percentages and the rest of the everyday arithmetic.

Common questions

Does the word counter upload the text I paste?

No. The counting happens in the page you already loaded: no request is made to our server, nothing is logged, and closing the tab discards the text. That is also why there is no usage limit — counting a paragraph once and counting it fifty times costs us exactly the same, which is nothing.

Why does your word count disagree with my editor?

Because "word" has no single definition. A counter that splits on whitespace reports a whole Chinese sentence as one word; one that splits on punctuation counts "don't" as two. This site writes its rules out — Chinese and Japanese count by character, other scripts by runs of letters and digits, apostrophes and hyphens join words — so a disagreement is something you can check instead of guessing at.

What is a character here — a byte, a code point, or what I can see?

A Unicode code point, so an emoji counts as one character rather than two and a flag is not four. Byte length is a different question with a different answer: the same Chinese sentence is about three times longer in UTF-8 than in UTF-16. When a form asks for characters it almost always means code points, and the page prints both so you can see which one you are over.

Will you add a plagiarism checker or an AI-text detector?

No. Both need a model and a corpus we do not have, and both hand you a probability dressed up as a verdict. A page that announces "83% written by AI" about a paragraph you wrote yourself is worse than no page at all. Text tools here do arithmetic on text: count it, change its case, remove the duplicate lines.

Which text tools are coming next?

Case conversion, line deduplication and sorting, text comparison, and a Markdown table builder. Every one of them is a pure string operation with no data behind it, which is exactly why they can be published honestly — and each ships in English and Chinese before it is listed here.