LaunchRanked

Readability checker

Paste your text. We score it five ways, show each formula with your numbers filled in, and highlight sentences that run long, likely passive voice and adverbs. It's for English text and runs in your browser.

0 words0 sentences

Your text stays in this browser tab. Nothing is sent to us.

Readability results

Paste some English text to see five readability scores, the formulas with your numbers in them, and the sentences that run long.

What the scores measure

Every formula uses the same two things: how long your sentences are and how long your words are. Flesch Reading Ease (1948) and the Flesch-Kincaid grade level (developed for the US Navy in 1975) count syllables. Gunning Fog (1952) and SMOG (1969) count words of three or more syllables, and SMOG needs a sample of 30 sentences. Coleman-Liau (1975) counts letters instead (formulas and history). None of them can tell whether a text is clear, accurate or useful.

What to aim for on a web page

Nielsen Norman Group usually recommends a 6th to 8th grade level for a general audience, and says 10th to 12th grade suits experts (NN/g). On Flesch Reading Ease, 60 to 70 is plain English. Don’t chase a number: a Portent study of 756,297 ranking pages found top-30 content averaging around the 11th grade, with no link between reading level and position (Portent).

Does Google use readability scores?

Google hasn’t said it does. In a January 2018 hangout, John Mueller said that as far as he knew Google has no basic algorithms that count words to work out a reading level (Search Engine Roundtable). Google’s SEO starter guide asks for text that is easy to read, well organized and free of spelling and grammar mistakes, and mentions no scores. That was one answer eight years ago, so it’s a data point, not a guarantee. Use the numbers to find hard sentences, not as a ranking target.

How we count

  • Words and sentences are split the same way as in the word counter. Lines wrapped in the middle of a sentence (text from a PDF or email) are joined. Headings and list items count as sentences, so paste body text for the cleanest score.
  • Links and email addresses aren’t words, and numbers count as one syllable. Acronyms in capitals, like SEO and API, are read letter by letter and left out of Fog’s complex words, which can put Flesch a point or two away from other tools.
  • Fog counts three-syllable words once -ing, -ed and -es are taken off, and skips names that don’t start a sentence. SMOG counts every word of three syllables or more.
  • We flag sentences over 25 words. That’s our line, not a rule: Oregon’s plain-language guide aims for 14 to 20 (Oregon DAS). Passive voice and -ly adverbs are heuristics: use them as prompts to look, not as errors.

Using it to edit

Split the longest sentences first, then swap long words for short ones where the meaning holds. For length and reading time use the word counter; to see which phrases a draft repeats, the keyword density checker; for titles, the headline analyzer.

Questions

Which readability score should I use?

Pick one and compare like with like. Flesch Reading Ease is the best known; the grade-level formulas (Flesch-Kincaid, Fog, SMOG, Coleman-Liau) are easier to read against an audience. They won't agree: on our own blog posts the four grades differed by about three from lowest to highest, with SMOG usually on top. Coleman-Liau uses letters, not syllables, so it makes a useful cross-check.

Is passive voice bad?

No. Passive voice is the right choice when the doer doesn't matter or isn't known: "The file was deleted." It reads weaker when it hides who did something, or when every sentence is built that way. We flag a form of "be" plus a past participle as likely passive. It's a heuristic, so it misses some cases and catches adjectives such as "is committed".

How accurate is the syllable count?

English spelling isn't regular, so counting syllables by rule is approximate. Checked against the CMU Pronouncing Dictionary, it matched about 98% of a sample of 2,000 common words (the list we tuned it on) and about 93% of 1,500 rarer words it hadn't seen. On about 50,000 words from our own blog posts, swapping in the dictionary's counts moved Flesch Reading Ease by 0.05 points on average and 0.2 at most.

Does it work for other languages?

No. The formulas were built for English and the syllable rules assume English spelling. If a text doesn't look like English prose, with few common English words or mostly non-Latin letters, the tool says so. Scores for other languages wouldn't mean much.

Why is my score different in another tool?

Tools differ in how they split sentences (do headings count?), count syllables, and treat numbers, acronyms and links. A difference of a point or two on Flesch is normal. Compare texts in the same tool, and don't mix scores from several.

AI SEO Autopilot

Want this fixed automatically?

Autopilot checks your site, publishes researched articles at a safe pace, and tracks whether ChatGPT, Perplexity, Gemini and Claude cite you. From $40.83/mo, billed yearly.

Related free tools

See all free tools