Flesch Reading Ease Explained
The Flesch Reading Ease score rates English text from 0 to 100, where higher means easier to read. It looks at only two things: how long your sentences are and how long your words are, measured in syllables. That makes it crude, but also fast, consistent and hard to argue with, which is why it has survived since 1948 and still ships inside Microsoft Word.
The short version
- Aim for 60 or above for web pages, emails and product copy.
- Below 30 reads like an academic paper or a contract.
- The two levers are shorter sentences and shorter words. Nothing else moves the score.
- It measures difficulty, not quality. A clear, well-organized page can score 50 and still be the best answer.
The formula
Rudolf Flesch published the formula in 1948. With every count taken from the passage you are scoring:
Reading Ease = 206.835 − 1.015 × (words ÷ sentences)
− 84.6 × (syllables ÷ words)The first term penalizes long sentences and the second penalizes long words. Word length carries far more weight: adding one syllable per word on average costs 84.6 points, while adding one word per sentence costs about one point. So a passage of short sentences can still score badly if it is full of words like “implementation” and “functionality”.
The formula is not actually capped. Very simple text can score above 100 and dense text can go below zero, so most tools, including ours, clamp the result to 0–100.
What the scores mean
Flesch tied score bands to the US school grade at which a reader could follow the text:
| Score | Difficulty | Roughly readable by |
|---|---|---|
| 90–100 | Very easy | 11-year-olds (5th grade) |
| 80–90 | Easy | 12-year-olds (6th grade) |
| 70–80 | Fairly easy | 13-year-olds (7th grade) |
| 60–70 | Plain English | 13–15-year-olds (8th–9th grade) |
| 50–60 | Fairly difficult | High school students |
| 30–50 | Difficult | College students |
| 0–30 | Very difficult | College graduates |
“Plain English” at 60–70 is the useful target for most writing aimed at the public. That is not because readers are children. Even expert readers move faster through simple prose, and people skim on screens.
A worked example
Here is a sentence written the way software documentation sometimes is:
Notwithstanding the aforementioned limitations, utilization of the application necessitates the installation of supplementary components prior to initialization.
It is one sentence of 17 words, and our Word Counter counts 51 syllables, or 3 per word. The formula gives 206.835 − 1.015 × 17 − 84.6 × 3 = −64, which shows as 0. Now the same instruction, rewritten:
The app has a few limits. Before you start it, install the extra parts it needs.
That is 16 words in 2 sentences, with 20 syllables. The score is 206.835 − 1.015 × 8 − 84.6 × 1.25 = 93. The meaning did not change; the words did.
Syllable counts are where tools disagree. Counting them exactly needs a pronunciation dictionary, so most tools, ours included, use a rule of thumb based on groups of vowels. For the sentence above the rule matches a careful hand count exactly, but on other text it can be off by a syllable here and there, as can every other tool’s rule. Expect scores from different tools to differ by a few points for the same text, and compare a draft against itself rather than against a number from another tool.
Flesch Reading Ease vs Flesch-Kincaid Grade Level
The Flesch-Kincaid Grade Level uses the same two inputs with different weights, and it outputs a US school grade instead of a 0–100 score:
Grade Level = 0.39 × (words ÷ sentences)
+ 11.8 × (syllables ÷ words) − 15.59It was developed in 1975 for the US Navy to grade training manuals. The two scales run in opposite directions: a higher Reading Ease is easier, while a higher Grade Level is harder. Our rewritten example lands at about grade 2, and the original at grade 26, which is beyond any real school grade. Because they share their inputs, the two always agree on whether a change made text easier. Pick one and stick with it.
How to raise your score
- Split long sentences. Look for “and”, “which” and semicolons joining two ideas. Each split shortens the average.
- Swap long words for short ones. “Use” for “utilize”, “start” for “initialize”, “help” for “facilitate”. This moves the score more than anything else.
- Turn nouns back into verbs. “Make a decision” becomes “decide”; “the installation of” becomes “install”.
- Keep the technical terms you need. If your readers search for “authentication”, write “authentication”, and make the words around it simpler instead.
Paste a draft into the Word Counter, which shows the score with words per sentence and syllables per word, and edit while you watch it change. Its keyword density list also points out long words you repeat.
What the score cannot tell you
The formula never looks at meaning. “The cat sat on the mat” and a jumbled list of short words score the same. It cannot tell a familiar long word like “everybody” from an obscure short one, and it says nothing about structure, headings or whether the reader finds the answer. Treat it as one signal for prose difficulty, not a grade for the writing.
It is also tuned for English. Other languages have longer average words, so unadjusted scores come out lower. Adapted versions exist, such as Amstad’s for German and Fernández Huerta’s for Spanish. The score in our Word Counter uses the original English formula.
Quick answers
What is a good Flesch Reading Ease score? 60–70 for general web content and email; 50 or higher is reasonable for specialist or technical writing.
Does Google use readability scores for ranking? Google has never said it uses the Flesch formula. Readable writing helps people understand and act on a page, which is what matters, but chasing a number is not a ranking strategy.
Why does my score differ between tools? Mostly syllable counting, and sometimes how sentences are split at abbreviations like “e.g.”. Compare drafts within the same tool.