Free tool · On-page SEO
Free readability checker
To check readability, measure average sentence length and average syllables per word, which is all the Flesch formulas use. Paste your text below to get Reading Ease and grade level alongside the three counts they are computed from, your longest sentences in full, and every passive clause shown as text you can overrule.
Free, and your text never leaves your browser. The extractability read of a published page needs a free sign-in.
The scores below are computed in your browser. Your text is not sent to us.
AEO depth layer
See the paragraph extractability read
The paragraph extractability read is computed from your own page and costs nothing to run. We ask for a free account because the expensive tools on this shelf stay free, and because it lets us tell you when your result changes. No card, no trial clock.
- Each paragraph of a live page judged on whether it could be lifted out and still make sense on its own.
- Paragraphs opening with It, This or They, which refer back to the one before and so cannot be quoted alone. This is invisible while you read top to bottom.
- Paragraphs too short to answer anything, and paragraphs too long to quote whole so an engine chooses the boundary instead of you.
Signing in saves the details you entered above so we can follow up about your results. See our privacy policy for what we keep and why. The Readability Checker result you already have stays free either way. Privacy policy.
The bands, and what each one costs you
The score ranges are the conventional Flesch bands. The descriptions are ours, and they are judgements rather than findings: a technical reference scoring 40 may be exactly right for its readers.
| Reading ease | Band | What it reads like |
|---|---|---|
| 90 and above | Very easy | Readable by an average 11 year old. Rare outside children's writing. |
| 70 to 89 | Easy | Conversational. Where most good marketing and documentation sits. |
| 60 to 69 | Plain | Plain English, readable by most adults without effort. A good target for a guide. |
| 50 to 59 | Fairly hard | Longer sentences and more multi-syllable words. Common in trade press. |
| 30 to 49 | Hard | Academic register. Fine for a specialist audience, costly for a general one. |
| below 30 | Very hard | Dense and clause-heavy. Almost always worth breaking up. |
Both formulas take only two inputs: average words per sentence and average syllables per word. That is the whole model, and it is why a long sentence of simple words and a short sentence of complicated ones can score the same. It measures surface difficulty, never whether an argument makes sense.
How this compares to a typical free checker
Where the usual checkers are equal, the table says so.
| Capability | This tool | Typical free checker |
|---|---|---|
| Flesch Reading Ease and grade level | Yes | Yes |
| Runs without sending your text anywhere | Yes | Usually posts it to a server |
| Shows the counts the score is built from | Yes | Score only |
| Sentence splitter guarded against decimals | Yes | Splits on 1.5 |
| Sentence splitter guarded against abbreviations | Yes | Splits on Fig. |
| Lists your actual long sentences | Yes, longest first | An average |
| Passive voice shown as clauses | Yes | A percentage |
| Ignores adjectives and idioms in passive detection | Yes | Flags them |
| States where its syllable heuristic fails | Yes | No |
| Paragraph extractability on a live page | Yes, with a free account | No |
What is in the paragraph extractability read
Readability tells you whether a person can follow your prose. This asks a different question: could a single paragraph be lifted out of it and still answer something? An answer engine quotes passages, not pages, and a paragraph that only makes sense in sequence is a paragraph it has to skip.
- Paragraphs opening with It, This, They, Therefore or However, which refer back to the paragraph before. Quoted alone they have no subject. This is the most common finding and it is invisible while you read top to bottom, because in sequence it reads perfectly.
- Paragraphs too short to answer anything, which are fine as transitions and useless as the passage an engine lifts.
- Paragraphs too long to quote whole, where an engine takes a fragment and picks the boundary itself. Split them and you pick.
A paragraph flagged as not citable is not badly written. It is written to be read in sequence, which is a different job and often the right one. Once the prose is extractable, the h1 checker reads whether your headings are shaped like the questions people ask, the AI citation schema generator marks up who wrote it and what it is based on, and the AI visibility checker shows whether the engines name you at all.
How to check the readability of your writing
Step 1
Paste the draft, not the published page
This runs in your browser and nothing is sent to us, so an unpublished draft is fine. Paste a section rather than a whole site; the scores mean more on one piece of writing.
Step 2
Read the counts before the score
If the sentence count looks wrong, the score is wrong, and everything downstream of it is too. That is why the counts are shown at all.
Step 3
Fix the longest sentences, then reread the passives
Sentence length is the input you control most directly. Passive voice is a choice: it is correct when the actor is unknown or irrelevant, so read each clause rather than removing them all.
Step 4
Check whether your paragraphs can stand alone
The extractability read judges a published page paragraph by paragraph. A paragraph opening with It or This cannot be quoted on its own, which matters when an answer engine is deciding what to lift.
When you outgrow this tool
Readable prose is the floor. Answer Radar tells you which questions your pages are actually cited for, and which competitor gets named instead.
The readability check is free and complete on its own, and runs entirely in your browser. Linkeddit is what you use when clear prose stops being the question and you need to know which of your pages the engines actually cite.
FAQ
Readability Checker questions
What the formulas actually measure, what they cannot tell you, and where this tool's heuristics are known to be wrong.
What is a good Flesch Reading Ease score?
For most business and marketing writing, 60 to 70 is a sensible target: plain English that an adult reads without effort. Above 70 reads as conversational, which suits a landing page. Below 50 is where a general reader starts working, and below 30 is academic register. None of these are rules. A technical reference for engineers scoring 40 may be exactly right for its audience, and a chatty blog post scoring 80 may still say nothing.
How is Flesch Reading Ease calculated?
206.835 minus 1.015 times the average words per sentence, minus 84.6 times the average syllables per word. Flesch-Kincaid grade level uses the same two inputs differently: 0.39 times words per sentence, plus 11.8 times syllables per word, minus 15.59. That is the whole formula, which is why this tool shows you all three counts. Two inputs drive everything, and both are things you can see and argue with.
Does readability affect SEO?
Not as a direct ranking factor, and anyone claiming a Flesch target will lift your rankings is guessing. What it affects is whether people finish reading and whether an answer engine can extract a clean passage from your page. Both matter, and both are downstream of clarity rather than of a number. Treat the score as a smoke alarm, not a scoreboard.
Is passive voice bad writing?
No. It is the right choice when the actor is unknown, irrelevant or obvious, and forcing every sentence into the active voice produces worse prose than leaving some alone. This tool shows you the clause rather than a percentage precisely so you can make that call. It also refuses to flag adjectives like "is interesting" and idioms like "was born", because a detector that cries wolf on those is one people learn to ignore.
How accurate is the syllable count?
It is a spelling heuristic, not a pronunciation dictionary, and it will be wrong on some words. A syllabic consonant such as in "organism" is undercounted, two vowels in separate syllables are read as one group, and names and loanwords follow rules no English heuristic captures. The tool lists these limits with example words. A real dictionary is megabytes we are not going to ship into a browser for a free tool, and saying so is better than implying precision we do not have.
What is in the layer that needs an account?
The extractability read. It fetches a published page and judges each paragraph on whether it could be lifted out and still make sense: paragraphs opening with It, This or They that refer back to the previous one, paragraphs too short to answer anything, and paragraphs too long to quote whole. That failure is invisible while you read top to bottom, which is what makes it worth a tool. It costs nothing to run and no model is involved.
How is this different from other free readability checkers?
It shows its working. Most tools return a score and a colour; this returns the sentence, word and syllable counts the score is built from, so a wrong split is visible rather than silently baked in. It also guards the sentence splitter against decimals and abbreviations, lists your actual long sentences instead of an average, and states where its syllable heuristic fails rather than presenting one decimal place as precision.
More free tools
Every tool on the shelf is free, with no card and no trial. Most need no account at all. See all free tools.
Check up to 25 descriptions at once and find the duplicates, which is the one problem you cannot see a page at a time.
See where a title and description get cut on desktop and mobile, measured in your own browser rather than estimated from a character count.
Word and phrase frequency with the denominator stated, plus an honest note that density is not a ranking factor.
Thirty-five checks worth a hundred points, with every weight and every threshold printed before you enter a URL.