Word & Character Counter
Count words, characters, sentences and paragraphs, with reading time, keyword frequency and byte size.
How to use this calculator
- 1Paste or type your text.
- 2Adjust the reading speed if your audience reads faster or slower than average.
- 3Check the average sentence length — it is a better readability signal than any single score.
How the calculation works
Words = tokens separated by whitespace Reading time = words / speed- Speed
- Reading rate in words per minute — 238 wpm is the average for silent reading of English prose
Sentence counting splits on full stops, question marks and exclamation marks followed by whitespace. Abbreviations such as "e.g." will inflate the count slightly.
Byte size is measured in UTF-8, where ASCII characters take one byte and most accented and non-Latin characters take two to four.
Worked example
A short two-paragraph passage
- 1.Whitespace splitting gives 8 words.
- 2.Two full stops give 2 sentences; the blank line gives 2 paragraphs.
- 3.At 238 words per minute, 8 words takes about 2 seconds.
Result: 8 words, 2 sentences, 2 paragraphs
What this tool measures, and why counts matter
Counting a piece of text sounds trivial, but the number that comes out — words, characters, sentences — is rarely just curiosity. It is almost always being checked against some limit or target set by someone else: a platform’s character cap, a style guide’s sentence-length guidance, an assignment’s word-count minimum, a search engine’s ideal snippet length. The count itself is simple arithmetic; what makes it useful is knowing which limit it needs to satisfy.
How the counts are actually produced
Word count here is produced by splitting the text on whitespace, which is the same basic approach word processors use — it treats a hyphenated compound like "well-known" as one word and handles most everyday English and other space-delimited languages correctly. Sentence count is heuristic rather than exact: it splits on a period, question mark or exclamation mark followed by whitespace, which works well for ordinary prose but can be thrown off by abbreviations ("e.g.", "Dr.") or decimal numbers, since those also contain a period followed by more text.
Character count and byte size answer two different questions. Character count is what most on-screen limits refer to — how many visible characters are in the text. Byte size, measured in UTF-8, is what actually gets transmitted or stored, and the two only match for plain ASCII text: an accented letter, a curly quote or an emoji can take two to four bytes for a single character.
Where a specific count becomes a hard limit
Character and word counts show up as real constraints across writing, publishing and development:
- Social platforms — a single post has always had a hard character cap — historically 280 on X/Twitter — and going over it either blocks posting outright or gets the text silently truncated.
- SEO meta descriptions and title tags — search engines truncate a page’s meta description around 155–160 characters and its title tag around 50–60, so anything longer just gets cut off with an ellipsis in the search results.
- Academic and professional writing — essays, cover letters and grant applications routinely specify a word-count range, and going meaningfully under or over it is often read as not following instructions.
- SMS and legacy messaging — a text message is still transmitted in 160-character segments; a message just over that limit doesn’t get rejected, it gets silently split into two.
Where counts get misread
A few mismatches trip people up regularly:
- Characters versus bytes — a limit stated in "characters" almost never means bytes, but some older systems and APIs do enforce a byte limit instead — which matters a lot once accented letters, emoji or non-Latin scripts are involved, since those can use several bytes per character.
- Words versus tokens — a word count is not the same as the "token" count language models and their APIs bill by — a token is often a sub-word chunk, so word count is only a rough proxy, not an exact substitute, when estimating usage against an LLM API.
- With spaces or without — some limits, like SMS, count every character including spaces; others have historically counted only non-space characters. Always check which one a given limit actually refers to before trimming text to fit.
What this assumes, and where it stops
Assumptions
- Words are whitespace-delimited, which suits English and most European languages.
Limitations
- Word counting does not work for languages written without spaces, such as Chinese, Japanese and Thai.
- Sentence detection is heuristic. Abbreviations, decimals and ellipses can be miscounted.
- Reading speed varies enormously with material and reader — technical text is read far more slowly than narrative prose.
Common questions
How is reading time calculated?
Word count divided by reading speed. The 238 words-per-minute default comes from a large meta-analysis of silent reading of English non-fiction. Dense technical material is read considerably slower, so adjust the speed if that is what you are measuring.
Does the character count include spaces?
Both figures are shown. Social platforms and SMS count everything including spaces; some publishing contracts count characters excluding them. Check which one your limit refers to.
Formula and content last reviewed on .
Results are estimates for information only, not professional advice.
Related calculators
Tools people commonly use alongside the word & character counter.