Text Statistics
Detailed analysis of your text — counts, averages, reading ease, and more.
Type or paste text above to see statistics.
How text statistics are calculated
Word count is Unicode-aware, so accented Latin, Cyrillic and Greek words are counted rather than discarded; character count includes every character, spaces included. Sentence count detects ., ! and ? while ignoring decimal points and common abbreviations, so 3.14 and e.g. do not split a sentence. It also detects the terminators other scripts use — the full-width 。 ！ ？ of Chinese and Japanese, the Devanagari danda ।, the Urdu ۔, the Arabic ؟ — and treats a hard line break as a boundary, so a bulleted list is counted as one sentence per item rather than one in total. Syllable count uses a vowel-group heuristic (accurate within ~5% for standard English prose), not a dictionary. The Flesch Reading Ease formula is 206.835 − 1.015×(words/sentences) − 84.6×(syllables/words) — scores near 100 are very easy to read; scores near 0 are academic/technical texts, and dense prose can score below zero, which is reported rather than rounded up.
For a quick word/character count, use the Word Counter. For reading time, see Reading Time Calculator.
Built and maintained by Meet Shah · Last updated
What this tool is used for
- Getting word, sentence and paragraph counts in one pass.
- Checking average sentence length against a readability target.
- Comparing two drafts on structure rather than impression.
- Finding whether a document's paragraphs are unusually long.
- Getting a Flesch score alongside the raw counts.
Frequently Asked Questions
- Which statistics are worth reading?
- Word and sentence counts as raw size, average sentence length as the strongest readability signal, and the count of distinct words against the total as a measure of repetition. The single number people over-read is a grade level; the average sentence length behind it is more actionable.
- How is a sentence counted?
- With a real splitter rather than by counting full stops — abbreviations, decimals and quoted terminators all produce false boundaries, and a naive count reads `Dr. Smith arrived.` as two sentences. It also handles CJK terminators, which have no trailing space to anchor on.
- Why does average sentence length matter so much?
- Because it is the variable every readability formula weights most heavily, and it is the one a writer can act on directly. Cutting a 40-word sentence in half improves comprehension measurably; swapping long words for short ones rarely does.
- What does lexical diversity tell me?
- How much the vocabulary repeats — distinct words divided by total. It falls naturally as a text gets longer, so comparing a paragraph against a chapter is meaningless. It is useful for spotting a document that leans on the same handful of phrases.
- Are the counts accurate for non-English text?
- Word counting is, because it segments scripts that do not put spaces between words rather than treating a Chinese paragraph as one token. Readability scores are not — they are fitted to English syllable and sentence patterns and carry no meaning elsewhere.
Common errors and gotchas
- Treating a readability score as a quality measure, when a low score can mean choppy prose.
- Applying English-tuned formulas to another language.
- Counting code blocks and captions as prose, which distorts every average.
- Splitting sentences on every period, which breaks abbreviations.
- Optimising the numbers rather than the writing.