Readability Analyzer

Score any page or draft on five readability formulas at once, and see which sentences and words are pulling the grade level up. Enter a URL to check a published page, or paste a draft before it goes live.

Content to score

Reading the score

Flesch Reading Ease runs from 0 to 100, and higher is easier. A score of 60 to 70 is plain English, comfortable for most adults reading on a screen. Below 50, the reader has to work, and on the web the reader usually declines. Above 80, the writing is simple enough for a general audience of any education level, which is the right target for anything a customer reads under time pressure.

The grade formulas answer a different question: how many years of schooling a reader needs to follow the text on the first pass. Five of them are shown because they disagree in informative ways. A page where Gunning Fog is far above Flesch-Kincaid is carrying a lot of long words; a page where the reverse holds is carrying long sentences of short words.

Sentence length is the lever with the most travel. Every formula on the page weighs it, and it is the one thing you can fix without changing what the sentence says. The longest sentences in your text are listed above for exactly that reason.

All five formulas were built for English, and each of them counts English syllables, English letters, or both. A draft in another language written with the Latin alphabet still gets a number, and it is a fair way to compare that draft against another one like it. Text in a script the formulas cannot read gets no number, because there would be nothing behind it.

Frequently Asked Questions

What is a good readability score?

For most web writing, a Flesch Reading Ease of 60 to 70 and a grade level around 8 or 9 is the range to aim for. That is not a dumbing-down target: it is roughly the level of a well-edited newspaper, and it is what lets a reader skim without losing the thread. Technical documentation written for specialists can sit lower without harm, because the audience already carries the vocabulary.

Is readability a Google ranking factor?

Not directly. Google has said it does not use readability scores as a signal. What it does measure is whether people find what they came for, and hard-to-read pages lose readers before they find it. So readability affects rankings the way page speed does: through the behavior it causes rather than through a score being read.

Why do the five formulas disagree?

Each one weighs different things. Flesch-Kincaid and SMOG count syllables, Coleman-Liau and Automated Readability count characters, and Gunning Fog counts words of three or more syllables. A page full of short technical terms scores easy on the syllable formulas and hard on the character ones. The average across all five is more stable than any single number, which is why it is shown separately.

How do I improve a readability score?

Split the long sentences first; sentence length dominates every formula on this page, and the longest sentences are listed above so you can find them. After that, replace the long words that have short equivalents, and break walls of text into shorter paragraphs. Cutting passive constructions helps the writing more than it helps the score.

Does this work for languages other than English?

The five formulas count English syllables and English letters, and every one of them was calibrated on English text. A draft in another language written with the Latin alphabet still scores, and two such drafts compare usefully against each other, but neither number lines up with an English one. Text written in another script is not scored at all. The counters find no syllable in Japanese, Russian, or Arabic, so the arithmetic would hand back a perfect 100 for any page written in them. The tool names the script and shows no number instead.

Can I check the readability of a live page?

Yes. Enter any URL and the tool fetches the page, takes the content out of it, and scores the writing. Navigation, buttons, code samples, and the one-line labels a page is built from are all left out, because a formula run over a menu describes nothing a reader reads. What comes back is the reading level of the prose rather than of the markup around it. This is the quickest way to compare your draft against the pages currently ranking for your query, which is a better target than any absolute score.