📝 Word Counter & Text Analyzer
Count words, characters, sentences, paragraphs. Keyword density analysis included.
Last updated: November 8, 2026 · By the BoltQuickTools team
Check length against a limit, and spot repeated words
Essays, abstracts, product descriptions and scripts for video all come with word or time limits. Paste a draft here to see how long it is, roughly how long it takes to read aloud, and which words you are leaning on too heavily.
Using it
Paste or type into Paste or type your text. The eight figures (Words, Characters, No Spaces, Sentences, Paragraphs, Reading Time, Speaking Time and Lines) and the Top Keywords (density) list update on every keystroke. Clear empties the box. Nothing else is needed.
Example, including where the counts go wrong
The invoice is late. Please send the invoice by Friday!
Dr. Lee paid 3.5 invoices.
| Figure | Result |
|---|---|
| Words | 15 |
| Characters | 83 |
| No Spaces | 68 |
| Sentences | 5 |
| Paragraphs | 2 |
| Reading Time | 1m |
| Speaking Time | 1m |
| Lines | 3 |
A reader would count three sentences. The page reports five because every run of ., ! or ? ends a sentence, so Dr. and the decimal point in 3.5 each add one.
The top keyword is invoice 2 (13.3%). Common words like the and is are filtered out. 3.5 appears as 35, and invoices is counted separately from invoice, since nothing is reduced to a root form.
How the figures are defined
- Words are anything separated by whitespace.
well-knownis one word, and a dash with spaces on both sides counts as a word, soword - wordis 3. - Languages without spaces are undercounted. Chinese or Japanese text between spaces counts as a single word.
- Characters counts JavaScript string length, so most emoji count as 2.
- Paragraphs are separated by a blank line, while single line breaks do not start one. A trailing newline adds to Lines and Characters but not to Words or Paragraphs.
- Reading Time is words divided by 238 and Speaking Time is words divided by 150, both rounded up to whole minutes. Even a one-word text shows 1m. Those are fixed averages and technical or spoken material varies a lot, so time a real read-through for anything important.
- Keyword density shows the top 20 words of two or more characters. Percentages use the total word count, including the filtered common words.
- Keywords work best in English. Characters outside a-z, 0-9, apostrophe and hyphen are stripped, so
caféis listed ascaf,naïveasnave, and Chinese words vanish from the list. The word and character totals are unaffected, and the common-word filter is English only.
Word processors, search rankings and prompts
Why doesn't this match the count in my word processor?
Word processors have their own rules for hyphens, dashes, footnotes and text boxes. Differences of a few words are normal. If a submission has a hard limit, use the count from the system that enforces it.
Does keyword density matter for search rankings?
Search engines do not publish a target percentage. It is more useful as an editing check for words you have overused, and a sign to rewrite when one term dominates a paragraph.
Can it check a language model's prompt length?
Words are not the unit models bill by. The LLM token counter estimates tokens instead.
Can I clean the text up from here?
This page only measures. Text tools removes empty lines and duplicates, and the case converter changes capitalisation. To compare two drafts, use the text diff checker.