DanfioSearch tools/

Type a tool name, a pinyin spelling or a synonym - both languages are searched.

Readability Score Checker

Global tools

Paste a draft and get the six readability scores a writer is usually asked about: Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and the Automated Readability Index - each with the one line of arithmetic it is, the band it lands in, and the grade level that goes with it. The scores are on the page before any JavaScript runs, because the analysis follows a fixed rule applied to the text and the built-in example runs the same rule.

Readability Score Checker - Danfio

https://danfio.com/tools/readability-checker/ - printed from this page

Everything happens in this tab as you type. There is no upload button on this page because there is no upload: the text never leaves the browser.

Whether `make` costs one syllable or two. It is the setting on this page that moves a Flesch score the most, and the table below shows both readings whatever you choose.

The word counter’s switch, doing the same work here: a longer average word lowers every score on the page.

Whether `2026`, `40%` and `3.14` are words. A report full of figures scores differently under each setting.

How many of the longest sentences to list. The table is capped by how many sentences the text has.

How many of the words the syllable rules are unsure about to name. Set it to 0 to hide the table entirely.

Leave it blank for "readability-report.txt" and "readability-sentences.csv".

Results update as you type. Nothing leaves your device.

Flesch Reading Ease

72.1

fairly easy - grade 7. 130 words in 7 sentences, 178 syllables.

Words
130
Sentences
7 18.6 words each
Syllables
178 136.9 per 100 words
Words of 3+ syllables
10 Fog and SMOG count these
Grade level
7.8 Flesch-Kincaid
Read time
39 sec 1 min read badge

The six scores for this text

ScoreThis textMeasured inReads as
Flesch Reading Ease72.10 to 100, higher is easierfairly easy - grade 7
Flesch-Kincaid grade7.8US school yearan 8th-grade reader, about age 13
Gunning Fog index10.5US school yearan 11th-grade reader, about age 16
SMOG index10.0US school yeara 10th-grade reader, about age 15
Coleman-Liau index7.8US school yearan 8th-grade reader, about age 13
Automated Readability Index8.1US school yearan 8th-grade reader, about age 13
Five of these answer in US school years, which is why they are printed next to each other rather than averaged: they are five fitted formulas, and a spread of two grades between them is normal.
Flesch Reading Ease is the only one on a scale of its own (0 to 100, higher is easier) and the only one a writer can quote without translating it first.

They disagree because they count different things: two count syllables, two count characters, and Gunning Fog counts only the words of three syllables or more.

The same text, with the final -e decided the other way

The ruleSyllablesSyllables per wordFlesch Reading Ease
A silent -e is dropped (make = 1) - this is the one above1781.36972.1
Every vowel group counts (make = 2)1951.50061.1
Counting the -e that English does not say makes a text look harder than it reads, and dropping an -e that a struggling reader really does meet makes it look easier.
This is the switch that explains most of the difference between two readability tools on ordinary prose, along with how each one counts a hyphen.

The word counter has the matching pair of switches for hyphens and numbers, and both pages count words with the same function, so a word count that disagrees cannot be blamed on this page.

The 5 longest sentences

#WordsSyllablesWords of 3+ syllablesSentence
627352If you disagree with the score, you can see which decisions produced it - a long sentence,…
425343The third is not, and no formula can make it so: a dictionary knows that business has two …
225342Each one is built from three numbers: how many words a sentence holds, how many syllables …
316190The first two are exact, because this page counts them the way the word counter does.
515190That is why the table below names the words this page is least sure about.
Sentence length is the one lever a writer can pull without reaching for a dictionary, and it is the first thing both Flesch and Fog punish.
The sentence number is its position in the text, so it can be found again after the text is split into paragraph rows.

Cutting the first row here in two usually moves the Flesch score more than replacing any single long word, and it helps a reader far more than it helps the number.

Words this page is least sure about

WordSyllablesDecided byTimes in this text
Readability5a letter pattern1
dictionary4a letter pattern1
decisions3a letter pattern1
business2the exception list1
every2the exception list1
there1the exception list1
Two things put a word here: it is in the exception list, where a dictionary says the rules would have got it wrong, or it holds a letter pattern such as `ea`, `io` or `ough`, where English decides the count word by word.
Those 6 words cover 6 of the 130 words in the text, so 4.6% of the score rests on the least certain part of the count.

A wrong syllable here moves the score by a fraction of a point; a sentence twice as long moves it by several. That is why the sentence table matters more, and why this one is printed anyway.

What a grade level means

GradeTypical ageWhat is usually written at that level
16-7a first reader: one short clause per sentence
38-9a children’s book
510-11middle grade fiction
7 - this text sits here12-13young adult fiction
914-15a newspaper report or a well-written blog post
1217-18a broadsheet editorial
1621+an academic paper or a contract
A US school year is a rough proxy for reading difficulty and a rougher one for age, but it is what these formulas return, so the page translates it into something a visitor can picture.
The row marked is where the Flesch-Kincaid grade of this text falls. The other grade-level scores above will usually mark a neighboring row, which is expected rather than a fault.

A grade level describes sentence length and word length, not how hard the idea is: a clear explanation of a difficult subject scores as easy as a nursery rhyme, and a legal paragraph about something simple scores as hard as a textbook.

The formula behind each score, and its one caveat

  • Flesch Reading Ease: 206.835 - 1.015 x words per sentence - 84.6 x syllables per word. This text averages 18.6 words and 136.9 syllables per 100 words, and the score lands in the "fairly easy" band: simple instructions and most first drafts of anything.
  • Flesch-Kincaid grade: 0.39 x words per sentence + 11.8 x syllables per word - 15.59. The same two ratios as the score above, fitted to a school grade instead of a 0-100 scale.
  • Gunning Fog index: 0.4 x (words per sentence + 100 x words of three syllables or more / words). This text holds 10 such words. The original index forgives -ed, -es and capitalized words; this one does not, so the fog number here runs a little higher than a tool that does.
  • SMOG index: 1.043 x the square root of (words of three syllables or more x 30 / sentences) + 3.1291. SMOG was fitted on 30-sentence samples and wants at least that many: this text has 7, so the further the sentence count is from 30, the less the number means.
  • Coleman-Liau index: 0.0588 x letters per 100 words - 0.296 x sentences per 100 words - 15.8. It counts characters and not syllables, so it is the one score here that does not lean on the least certain part of this page.
  • Automated Readability Index: 4.71 x characters per word + 0.5 x words per sentence - 21.43, characters counting letters and digits. It was built for the US Air Force to check training manuals, and it is the one that punishes a long word without asking whether it is a hard one.

The whole answer, as a report

Readability report

The numbers the scores are built from
  Words........................... 130
  Sentences....................... 7
  Syllables....................... 178
  Words of 3+ syllables........... 10
  Words of 7+ letters............. 20
  Letters......................... 558
  Average sentence................ 18.6 words
  Syllables per 100 words......... 136.9

Scores
  Flesch Reading Ease............. 72.1 out of 100
    fairly easy - grade 7
  Flesch-Kincaid grade............ 7.8 grade level
    an 8th-grade reader, about age 13
  Gunning Fog index............... 10.5 grade level
    an 11th-grade reader, about age 16
  SMOG index...................... 10.0 grade level
    a 10th-grade reader, about age 15
  Coleman-Liau index.............. 7.8 grade level
    an 8th-grade reader, about age 13
  Automated Readability Index..... 8.1 grade level
    an 8th-grade reader, about age 13

Rules in force
  Hyphenated compounds............ split into their parts (well-known = 2)
  Standalone numbers.............. counted as words
  Final -e........................ silent and not counted (make = 1)
  Syllable counting............... rules, with a list of words the rules are known to get wrong
  Flesch band..................... fairly easy - grade 7

The 5 longest sentences
     6.   27 words,   35 syllables, 2 of 3+ syllables
        If you disagree with the score, you can see which decisions produced it - a long sentence, a well-known compound, or a c…
     4.   25 words,   34 syllables, 3 of 3+ syllables
        The third is not, and no formula can make it so: a dictionary knows that business has two syllables, and a rule can only…
     2.   25 words,   34 syllables, 2 of 3+ syllables
        Each one is built from three numbers: how many words a sentence holds, how many syllables a word holds, and how many sen…
     3.   16 words,   19 syllables, 0 of 3+ syllables
        The first two are exact, because this page counts them the way the word counter does.
     5.   15 words,   19 syllables, 0 of 3+ syllables
        That is why the table below names the words this page is least sure about.

Words the count is least sure about
    5 Readability          letter pattern x1
    4 dictionary           letter pattern x1
    3 decisions            letter pattern x1
    2 business             dictionary x1
    2 every                dictionary x1
    1 there                dictionary x1

Nothing was uploaded: the text never left the page that scored it.

Copy it, or download it. The bytes in the box are the bytes in the file, and the first download button hands you exactly this.

What the numbers do not say

  • Syllables are counted by rules rather than looked up: a run of vowel letters is one syllable, a consonant before a final `-le` keeps that `-e`, a final `-e` is silent unless it is `ee` or `oe`, and a final `-ed` only counts when the `e` is spoken. On top of the rules sits a list of 51 words a dictionary says the rules get wrong, and the table above names every word this text put in doubt.
  • The final -e is the largest single lever here: dropping it gives 178 syllables and a Flesch score of 72.1, counting it gives 195 syllables and 61.1. Both readings are defensible, which is why it is a switch rather than a hidden decision.
  • Words and sentences come from the same function the word counter uses, so the hyphen and number switches here behave exactly as they do there. A word count that disagreed between the two pages would be impossible; a syllable count that disagrees with another tool is expected.
  • SMOG was fitted on thirty-sentence samples and this text has 7, so its number is an extrapolation - which is why it is printed beside four other scores rather than on its own.
  • A score describes sentences and words, not ideas. A clear explanation of a difficult subject scores easy, and a vague sentence about something trivial scores hard, so the number describes the writing and is never a verdict on it.
  • The formulas are English. Syllable counting is English letter patterns and the scores were fitted on English schoolbooks; for Chinese or Japanese the shared counter counts each character as a word and these syllable rules mean nothing at all.
  • The grade levels were fitted in 1948 (Flesch), 1969 (SMOG) and 1975 (Flesch-Kincaid) on the school textbooks of their day. A technical manual is not a school textbook, and a newsletter for people who already know the subject should not be written for grade 8.
  • Only the text in the box is scored: headings, bullet lists, tables, captions and link text are invisible here, so a page that reads well because of its layout scores worse than it reads.
Scored in this tab by the same function that produced the numbers already in the page you loaded, so nothing was sent anywhere, and there is no endpoint behind this page that could accept a draft even if something tried to send one.

What makes this page different from the other hundred readability checkers is the syllable count, which is the only part of the arithmetic that cannot be looked up. English does not say how many syllables "business" has; a dictionary does. So the rules that count them are printed here, the words the rules are known to get wrong are in a list, and the words this page was unsure about are shown by name - which means a score you disagree with can be traced back to the decision that caused it instead of being taken on trust.

About this readability score checker

Syllables are the one part of a readability score that cannot be looked up, and they are where every checker quietly disagrees. This page computes six scores - Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and ARI - and shows the sentences, the syllable rules and the words it was unsure about behind each number.

Nothing is uploaded: the analysis follows a fixed rule applied to the text, run in the built-in example and again in your browser, so a manuscript or an internal memo never leaves your machine. Because the syllable rules and their known exceptions are printed by name, a score you disagree with can be traced to the decision that produced it instead of being accepted on faith.

A readability grade is a description of sentence and word length, not a verdict on whether the writing is any good, and the page draws that line clearly. There is no account and nothing to sign up for; read the six scores as six opinions about difficulty, and read the flagged sentences as a to-do list rather than a score to beat.

How this readability score checker works

Counting syllables, which is where the honesty has to go

Four rules count the syllables: every run of vowel letters is one, a consonant before a final `-le` keeps that `-e` so `table` is two, a final `-e` is silent unless it is `ee` or `oe` so `make` is one, and a final `-ed` only counts when the `e` is spoken so `walked` is one while `wanted` is two. On top of the rules sits a short list of words a dictionary says they get wrong - `business` (two), `people` (two), `queue` (one), `readability` (five), `unfortunately` (five) - each one added after the rules were run over a paragraph and the counts checked by hand.

That is not a dictionary and the page does not pretend it is. What it does instead is flag the words where a dictionary might well disagree: the ones the exception list decided, and the ones holding a letter pattern where English settles the count word by word - `ea` is one syllable in `read` and two in `create`, `io` is one in `nation` and two in `radio`, and `ough` is anything it likes. Those words are listed under the score with their counts and how often they appear, so the least certain part of a Flesch score is the part a visitor can check.

Six formulas, because they disagree, and the disagreement is the useful part

Flesch Reading Ease (1948) is the only one on a 0-100 scale, and it is 206.835 minus 1.015 times the words per sentence minus 84.6 times the syllables per word: two ratios, fitted to a reading test, where a higher number is easier. Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and the Automated Readability Index all answer in US school years, and they disagree with each other by a grade or two on the same text, because two of them count syllables, two count characters, and Fog counts only the words of three syllables or more.

This page prints all six in one table, each with its formula and its own caveat, rather than picking one and calling it the answer. The Fog row admits that the original index forgives `-ed`, `-es` and capitalized words while this page does not, so its number runs a little higher than a tool that does. The SMOG row admits that the index was fitted on thirty-sentence samples - so on a five-sentence paragraph its answer is an extrapolation, and the row says so.

The three switches, and why they matter more than the last decimal place

The largest lever on a Flesch score is not a word, it is a decision: whether the silent final `-e` costs a syllable. Dropping it is what a dictionary does and what most tools do; counting it is what a struggling reader actually meets on the page. Both readings are shown in a table whatever the switch is set to, so the size of the disagreement is visible rather than argued about - and on ordinary prose it moves the score by several points.

The other two switches - whether `state-of-the-art` is one word or four, and whether `2026` is a word at all - are the word counter's, and they are here because the words and sentences in this analysis come from that tool's counting. Both pages count words with the same rule, so a word count that disagreed between them is impossible; a syllable count that disagrees with another readability tool is expected, and this page is where you find out which decision caused it.

What a score describes, and the four things it does not

A readability score describes sentences and words, not ideas. A clear explanation of a difficult subject scores as easy as a nursery rhyme, and a vague sentence about something trivial scores as hard as a textbook - which is why the page prints the longest sentences with their word and syllable counts, so the advice a score turns into is visible instead of implied.

It also only sees what you paste. Headings, bullet lists, tables, captions and link text are invisible to it, so a page that reads well because of its layout scores worse than it reads. The formulas are English, and were fitted on the schoolbooks of 1948, 1969 and 1975: a technical manual is not a school textbook. And the grade levels are a US convention that translates awkwardly everywhere else, so the table explaining what each grade means is printed beside them.

Worked examples

Each example below was run through the calculator on this page when the site was built, so the numbers match what you see when you enter the same inputs.

The same draft, scored with the silent final -e counted

A silent final -e: Counted (make = 2 syllables)

Flesch Reading Ease

61.1

plain English - grades 8-9. 130 words in 7 sentences, 195 syllables.

Words: 130 · Sentences: 7 · Syllables: 195 · Words of 3+ syllables: 14

Everything else is the form as it opens, so this is the sample text under one changed setting: counting the `-e` that English does not say costs 17 more syllables and takes the Flesch score from 72.1 down to 61.1, which is the difference between "fairly easy" and "plain English". It is the clearest example of why two readability tools disagree, and the page prints both readings side by side whatever this switch says.

A draft too heavy for a public page, and the sentences to blame

The text to score: The implementation of institutional accountability mechanisms necessitates comprehensive documentation of administrative procedures. Stakeholder engagement is therefore a prerequisite for the successful realization of the aforementioned objectives. · Rows in the sentence table: 3 · Rows in the unsure-words table: 5

Flesch Reading Ease

-73.2

very confusing - college graduate. 26 words in 2 sentences, 82 syllables.

Words: 26 · Sentences: 2 · Syllables: 82 · Words of 3+ syllables: 17

The longest-sentences table is the part a writer can act on, and this pair is the reason the page prints it: the two sentences are 12 and 14 words long, hold 45 and 37 syllables between them, and nine and eight of their words have three syllables or more. Every score lands between 25 and 32 on a school-year scale, and the Flesch score is negative - which is not a bug but the formula saying that no ordinary reader could follow this paragraph. Splitting either sentence moves the score further than replacing any single word.

A phrase the page refuses to score, and what it prints instead

The text to score: Send it now.

Too short to score

3 words

A score needs at least ten words and at least one sentence: this text holds 1 sentence and 3 syllables.

Words: 3 · Sentences: 1 · Syllables: 3 · Syllables per word: 1.000

Ten words in at least one sentence is where this arithmetic starts to mean something. Below that the page refuses, because one short sentence makes the average sentence length an accident of punctuation - and it prints the counts it did manage, including the syllables per word, rather than a score it would have to apologize for.

Frequently asked questions

What is a good Flesch Reading Ease score?

For a general audience, 60 to 70 is the band to aim at, and it is labeled "plain English" here: that is where news writing and good instructions sit. Above 80 is easy, and below 50 is difficult enough that most readers who did not choose the topic will stop. The band table under your score names the band, the school year that goes with it, and what the band means in practice, so the number can be read without a lookup table.

Why does this score disagree with the one I got somewhere else?

Almost always one of three things: syllables (whether a silent `-e` is counted, and how compounds are split), sentences (whether an abbreviation or a decimal point ends a sentence), or the version of the formula - this page counts every three-syllable word for Gunning Fog, and a tool that forgives `-ed` and `-es` will report a lower fog index. The switches above let you reproduce any of those choices, and the table showing both readings of the final `-e` says exactly how many points each one is worth.

How accurate is the syllable count?

Right on the ordinary words and wrong on a known list, which is why the list and the uncertain words are both printed. English has no rule that decides how many syllables "every", "fire" or "queue" has - only a dictionary knows - so a tool that does not admit this is either carrying a large pronunciation dictionary or hiding a guess. This page names every word it was unsure about, with the count it used and how often the word appears, so the uncertainty on your text is visible instead of claimed to be zero.

Is a lower grade level always better?

No, and a page that says so is lying to you. A grade 6 target is right for a landing page, a public notice and a help center article, and wrong for a contract, a clinical guideline or a newsletter for specialists - readers who chose the text will be insulted by an article that reads as though they had not. What the scores are good for is finding accidents: the 48-word sentence, the paragraph where four long words landed in a row, the draft that got harder to read after an edit.

Which of the six scores should I use?

Flesch Reading Ease for a general audience, because it is the one on a scale people recognize and the one your colleagues will have heard of. Flesch-Kincaid when a school grade is what the brief asks for. Coleman-Liau and the Automated Readability Index when the text is full of numbers, initials and short technical tokens, because they count characters rather than syllables and therefore do not depend on the least certain part of the count. Editing one sentence and watching all six move is the useful exercise here: if the six move together, you have found something real.

Does it work for languages other than English?

Not honestly, and the page says so when it notices. The syllable rules are English letter patterns and the formulas were fitted on English schoolbooks, so on a French or German text the numbers are arithmetic on the wrong language. Chinese and Japanese are worse: they are written without spaces, so the shared counter counts each character as a word and a syllable-based score becomes meaningless - when more than a third of the words are Han or Kana this page prints a warning instead of quietly reporting a number.

Is my text uploaded?

No. The whole analysis follows a fixed rule applied to the string in the box, so it runs in your browser and nothing leaves the tab: there is no upload button because there is no upload, and no endpoint behind the page that could accept a draft even if something tried to send one. That matters more here than on most pages, because the text people paste into a readability checker is usually an unpublished draft - a pitch, a performance review, a legal paragraph that is not ready for anybody else to read.

Why is the report in a box as well as in a table?

Because the report is the artifact people keep. It is the same six scores written out as plain text with the rules that produced them, the longest sentences, and the words the count was unsure about - and it is the first of the two downloads, so the numbers quoted in a review are the numbers the file holds rather than a retyped copy of them. The second download is the sentence table as CSV, for anyone tracking sentence length across drafts in a spreadsheet.

What is the shortest text this page will score?

Ten words in at least one sentence, and below that it refuses rather than printing a number it cannot defend: one short sentence makes the average sentence length an accident of punctuation, and SMOG's square root of a fraction is not a grade level. The refusal still shows the words, sentences and syllables it counted, and the syllables-per-word ratio, because those are real at any length.

Assumptions and sources

  • The six scores are the published formulas: Flesch Reading Ease and Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and the Automated Readability Index.
  • Syllable counting follows the page's own stated rules and exception list; a word with no clear syllable break is marked uncertain rather than guessed at.
  • Scores are a reading aid, not a verdict; the text is analyzed in your browser and never uploaded.

Last reviewed 2026-09-17. Runs locally on your device; keep backups; no financial advice applies. Read the full disclaimer.

Where a figure on this page comes from an official agency — HUD, IRS, FICO or a state body — the site keeps a dated snapshot of the source page in its sources library, and a scheduled job re-checks that snapshot, so the number can be traced back to the document it came from.

Related tools

Word & Character CounterText & WritingCount words, characters, sentences and paragraphs as you type: reading time, word frequency, and the two rules that make counters disagree. Nothing is uploaded.Count words and characters →Case Converter & Text CleanerText & WritingConvert text to any of twelve cases - Title Case, UPPERCASE, camelCase, snake_case and more - and clean what a paste brought along: odd spaces, curly quotes.Change text case →Text Diff & CompareText & WritingCompare two texts line by line or word by word: a unified diff, a change list with the words that moved, and switches for case, spacing and blank lines.Diff two texts →Lorem Ipsum GeneratorText & WritingGenerate lorem ipsum by paragraph, sentence or word - with a seed that gives the same text back every time, HTML or plain text, and the exact character count.Generate lorem ipsum →Acronym GeneratorText & WritingTurn a phrase into an acronym of first letters, with a letter-by-letter table of where each initial came from. Deterministic, private, honest about trademarks.Make an acronym →Business Name GeneratorText & WritingTurn a few keywords into business name ideas, usernames and hashtags. Six naming shapes, seeded so the list reproduces, and no false claims about availability.Generate business names →