Word & Character Counter

Word and character counter — live text statistics

This tool gives you instant text statistics as you type: word count, character count (with and without spaces), sentence count, paragraph count, and estimated reading time. It is built for writers, students, marketers and anyone working against a length limit — checking an essay against a word target, keeping a tweet or meta description under a character cap, or estimating how long a script will take to read aloud.

The counting rules, exactly

The text is trimmed and split on whitespace to count words. Characters with spaces is the full string length; characters without spaces strips all whitespace first. Sentences are found by splitting on ., ! and ? and discarding empty fragments. Paragraphs are blocks separated by a blank line. Reading time divides the word count by 200 words per minute and rounds up to the nearest second — 200 wpm is a standard benchmark for adult silent reading.

Example

Paste a 1,000-word blog post and you will see roughly 1,000 words, about 5,800–6,200 characters with spaces, the exact sentence and paragraph counts, and a reading time of 5 minutes (1000 ÷ 200). Trim it to 750 words and the reading time drops to about 3 min 45s.

When to use it

Reach for this tool any time a length limit governs what you can publish. Social posts are the clearest case: a tweet caps at 280 characters, a LinkedIn post truncates with a “see more” fold around 210 characters, and an Instagram caption allows 2,200. Search snippets are another: a meta description that runs past roughly 155–160 characters gets cut with an ellipsis in Google results, so the “characters with spaces” tile is the figure to watch. Students use it to hit an essay word target without padding; speakers use the reading-time estimate to size a script to a time slot; translators and copywriters use the character count to price work that is billed per character or per word.

Reading the six tiles

Each tile answers a different question. Words is what assignment briefs and content guidelines usually mean. Characters with spaces is the raw length and is what most strict platform limits count. Characters without spaces is useful when a limit explicitly excludes whitespace, or when comparing the density of two drafts. Sentences and paragraphs hint at readability — very long sentences or wall-of-text paragraphs show up as a low sentence count against a high word count. Reading time turns the word count into a human figure your audience actually feels.

Tips and common mistakes

The most common surprise is the sentence count. Because sentences are detected by splitting on ., ! and ?, an abbreviation such as “e.g.” or “U.S.A.” adds phantom sentence boundaries and inflates the count slightly. Decimal numbers like “3.5” can do the same. Treat the sentence figure as a guide, not an exact parse.

A second trap is character counting for social platforms. Some emoji and many flag or skin-tone variants are stored as more than one code unit, so a single visible emoji can register as two or more characters — which is exactly how the platform itself will count it against your limit, so the number here is the one you want. Finally, remember that spoken delivery is slower than silent reading: the reading-time tile uses 200 words per minute for silent reading, but a speech read aloud lands closer to 130–150 words per minute, so multiply by roughly 1.3 to estimate stage time.

Everything runs entirely in your browser — your text is never uploaded or stored.

Common length limits, for calibration

The counts become actionable when set against the limits you’re writing for: a standard post on X caps at 280 characters, an SMS segment at 160 GSM characters (70 for Unicode text), meta descriptions display well below about 160 characters, and Google Ads headlines cut at 30. Academic and publishing word limits (abstracts, personal statements, grant sections) are usually hard caps enforced by submission systems that count words much as this tool does — but when a limit matters, paste into the destination system early, because tokenisation quirks (hyphenated compounds, slashes, em-dashes) can shift counts by a few words between counters.

What counts as one “character” is a standard, not an opinion

Emoji, accented letters and non-Latin scripts expose a subtlety: what users perceive as one character can be several Unicode code points (a family emoji is up to seven, joined invisibly). The rules for segmenting text into user-perceived characters and words are standardised in Unicode Standard Annex #29, and browsers expose them through Intl.Segmenter. Different tools legitimately count differently depending on whether they count code points, UTF-16 units or grapheme clusters — which is why a tweet-length checker and a database length constraint can disagree about the same string, and why the definitive count for any limit is the one enforced by the destination system.