Skip to content
ToolBoxGenie

Readability Score Checker

Text Tools · Added 18 August 2026

Paste a paragraph or a whole draft and get six readability formulas at once: Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and the Automated Readability Index. Showing them together is the point — they disagree by a grade or two on ordinary prose, and the spread between them tells you more than any single figure. The counts each one is built from are listed underneath, along with the longest sentences, which is where a rewrite actually pays.

Scores settle down at around 100 words. Below 30 they move a lot on a single sentence.

How to use the readability score checker

  1. 1Paste your text into the box. Around 100 words is enough for the scores to settle; below 30 they swing on a single sentence.
  2. 2Read the Flesch Reading Ease score and its band — 60 to 70 is the range usually described as plain English.
  3. 3Compare the five grade-level formulas. Close agreement means the reading is straightforward; a wide spread usually means unusual vocabulary or uneven sentence lengths.
  4. 4Check the counts underneath to see what drove the result: average sentence length and the proportion of long words do most of the work.
  5. 5Work through the longest sentences listed at the bottom. Splitting those moves every score more than swapping individual words does.

Examples

Plain writing

Input
The rain in Spain falls mainly on the plain. It arrives in short bursts, and it leaves as quickly as it came.
Result
Reading Ease around 85, consensus grade around 4 - easy

Short sentences and one-syllable words. The formulas agree closely because there is nothing unusual for them to disagree about.

Institutional prose

Input
Notwithstanding the aforementioned meteorological considerations, the accumulated precipitation levels documented throughout the region demonstrate substantial interannual variability.
Result
Reading Ease near zero, consensus grade well past 17

One 22-word sentence of mostly four-syllable words. Every formula is dominated by those two facts.

The same idea, rewritten

Input
Rainfall in the region varies a lot from year to year.
Result
Reading Ease around 80, consensus grade around 5

Same meaning, half the words. This is what the scores are actually measuring.

About the readability score checker

What these formulas actually measure

All six reduce to two things: how long the sentences are, and how long or unusual the words are. Flesch Reading Ease weights syllables per word roughly five times as heavily as it weights sentence length, which is why replacing long words has an outsized effect on it. The grade-level formulas balance the two more evenly, so on those, sentence length does most of the work.

Nothing in any of them examines meaning, structure or argument. A formula cannot tell that a paragraph buries its point, that a term is undefined, or that two sentences contradict each other — and those are usually the real reasons something is hard to read. The score is a smoke alarm, not an inspection.

Where the scores came from

Rudolf Flesch published the Reading Ease formula in 1948, and J. Peter Kincaid adapted it for the US Navy in 1975 to grade training manuals, which is why the output is a school grade. Robert Gunning developed the Fog index for newspapers in 1952, and G. Harry McLaughlin published SMOG in 1969 specifically because he thought the existing formulas underestimated difficulty for health material.

That history explains their quirks. They were built for prose in a particular register and validated against schoolchildren of the period, so they behave oddly on anything unlike it: bulleted lists, code, headings, tables of figures and dialogue all produce numbers that are technically correct and not very meaningful. Score the running prose, not the whole page.

Using a score to actually improve a draft

Start with the longest sentences rather than the vocabulary. Every formula here divides by the sentence count, so splitting one 45-word sentence in three moves the result further than swapping a dozen long words does — and it usually helps the reader more, because length is what forces someone to hold several ideas at once.

Then look at the words that are long without being precise. Technical terms your audience knows are fine and will drag the score down anyway; that is a limitation of the formula, not a fault in the writing. What is worth changing is the register that adds syllables without adding meaning — utilise for use, in order to for to, at this point in time for now.

Finally, ignore small movements. A difference of half a grade is inside the noise of the syllable estimate and the sentence splitting. A move from grade 16 to grade 10 is real; a move from 9.2 to 8.8 is not worth a second draft.

Frequently asked questions

Which score should I use?
Flesch Reading Ease if you want one number and a band, because it is the most widely recognised and several style guides and insurance regulators are written against it. Flesch-Kincaid if you need a school grade level, since that is what most institutional requirements name. SMOG if the material is about health, where it is the established choice. In practice the consensus figure here — the median of the five grade formulas — is more robust than any one of them.
What is a good score?
For public-facing writing, a Reading Ease of 60 to 70 and a grade level of 8 or 9 is the usual target, which is roughly where a broadsheet newspaper sits. Technical documentation aimed at specialists can reasonably be higher. What matters more than hitting a number is that the score matches the audience: a grade 6 target makes sense for public health material and none at all for a research paper.
Why do the formulas disagree with each other?
Because they measure different proxies. Flesch-Kincaid and SMOG count syllables, Coleman-Liau and ARI count letters, and Gunning Fog counts words of three or more syllables as a share of the whole. A text full of long but familiar words scores badly on the syllable-based formulas and better on the letter-based ones. When the spread is wide, that difference is itself the finding.
How accurate is the syllable count?
It is an estimate, as it is in every implementation of these formulas. English spelling does not mark syllable boundaries, so short of a pronunciation dictionary the count comes from rules about vowel groups and word endings. This tool handles the cases that shift results on ordinary prose — silent final e, the difference between washed and wanted, table and makes — and unusual words and proper nouns will still be off by one here and there. The effect on the final grade is a fraction of a level rather than a whole one.
Does a good score mean the writing is good?
No. None of these formulas reads the text. They count words, sentences, letters and syllables, so a passage of fluent nonsense scores the same as the sentence it was modelled on, and text can be chopped into short sentences until the score improves while becoming harder to follow. Use the score to find dense passages, then judge the rewrite yourself.
How does this differ from a word counter?
A word counter tells you how much there is: words, characters, paragraphs, reading time. This tool tells you how hard it is to read, which is a different question with a different answer — two pieces of the same length can be several grade levels apart. The counts here exist to explain the scores rather than as the result.

Related tools