Flesch Reading Ease and Grade Level, Explained

Flesch Reading Ease and the Flesch-Kincaid grade level are two formulas that turn the same two measurements, average sentence length and average syllables per word, into a difficulty score. Reading Ease runs roughly from 0 to 100, where higher means easier. The grade level gives the same judgement as a United States school grade, where lower means easier. The Reading Time Calculator works out both for any text you paste, alongside the reading time.

Neither score knows what your text means. They count, and that is useful as long as you know what is being counted. This guide shows the formulas, a paragraph rewritten with both scores measured before and after, and the places where the counting goes wrong.

The two formulas

Rudolf Flesch published the Reading Ease formula in 1948. In 1975 J. Peter Kincaid and colleagues adapted it, for the United States Navy, into a formula that outputs a school grade. The calculator uses the standard versions of both:

• Flesch Reading Ease = 206.835 minus 1.015 times (words per sentence) minus 84.6 times (syllables per word)
• Flesch-Kincaid grade level = 0.39 times (words per sentence) plus 11.8 times (syllables per word) minus 15.59

Both get worse as sentences get longer and as words get longer. Syllables per word carries the heavier weight in each: one extra syllable per word, on average, costs 84.6 points of Reading Ease, while one extra word per sentence costs about one point. That is why replacing long words often moves the score more than splitting sentences does.

The calculator labels Reading Ease in plain terms: 80 and above is "Very easy", 60 to 79.9 "Plain English", 50 to 59.9 "Fairly hard", 30 to 49.9 "Difficult" and below 30 "Very difficult". Many writing guides suggest aiming for 60 or more for general web copy. That is a rule of thumb, not a standard, and not a search ranking factor.

One paragraph, before and after

Here is a paragraph written the way a lot of advice about titles is written:

In order to facilitate the optimisation of organic visibility, it is imperative that practitioners give consideration to the comprehensive alignment of title elements with the informational requirements of prospective visitors, particularly in circumstances where search engines may subsequently substitute alternative formulations. Consequently, an iterative methodology incorporating systematic measurement of character and pixel dimensions is recommended prior to publication.

And here is the same advice, rewritten:

Write a title that tells searchers what the page answers. Search engines may rewrite a title that does not, so say it plainly. Before you publish, check the length. Measure it in pixels as well as characters, then fix anything that runs long.

Measure	Before	After
Words	58	43
Sentences	2	4
Syllables	149	58
Words per sentence	29	10.8
Syllables per word	2.6	1.3
Flesch Reading Ease	-39.9	81.8
Flesch-Kincaid grade	26	4.5

The rewrite keeps every instruction: match the title to what the reader wants, expect search engines to rewrite titles that do not, measure before publishing, in pixels as well as characters. It loses the padding ("in order to facilitate", "it is imperative that practitioners give consideration to") and swaps long abstract words for short concrete ones. The score moved by more than 120 points, and the paragraph is easier to read. In this case the number and the judgement agree. That will not always be true.

Notice the "before" score is below zero and its grade level is 26. The formulas are not capped, so very dense text runs off the bottom of the scale and very simple text off the top. "The cat sat on the mat." scores 116.1, with a grade of -1.4. Treat anything outside 0 to 100 as "extremely hard" or "extremely easy" rather than a precise reading.

How the calculator counts syllables, and where it slips

Nobody counts syllables by dictionary at this speed, so the calculator uses a standard rule of thumb: drop a silent "e" or an "-es" or "-ed" ending, then count groups of vowels. It is right for most common words and wrong for some. These are the calculator's counts for single words:

Word	Calculator's count	Usual dictionary count
readability	5	5
people	2	2
business	3	2 or 3, depending on pronunciation
create	1	2
going	1	2
idea	2	3
recipe	2	3
queue	2	1
hyphenated	3	4

Most of the misses undercount, which nudges the score slightly in the easy direction. Over a paragraph or more the errors mostly wash out, which is why the score is worth reading for a whole draft and not worth much for a single sentence.

Sentence counting matters as much

The calculator ends a sentence at every full stop, question mark or exclamation mark. Abbreviations confuse it. Compare these two versions of one sentence:

Text	Sentences counted	Reading Ease	Grade
Titles are cut at about 600 px, e.g. on a desktop screen, and descriptions at about 920 px.	3	98	1.2
Titles are cut at about 600 px, for example on a desktop screen, and descriptions at about 920 px.	1	76.2	7.3

The "e.g." splits one sentence into three, which more than halves the words per sentence and lifts the score by more than 20 points. Nothing about the sentence became easier. Semicolons go the other way: "Check the width; fix the title; publish" counts as a single sentence. If your text uses a lot of abbreviations, decimals or semicolons, read the sentence count before trusting the score.

What the scores cannot tell you

• Whether the text makes sense. Shuffle the words of an easy paragraph and the score stays the same.
• Whether the words are familiar. "Cat" and "nub" score alike; so do "readability" and a five-syllable word nobody knows.
• Whether the order of ideas helps. Structure, headings, examples and the answer coming first all matter a great deal to readers, and none of it is measured.
• Your audience. A grade 12 score is fine for a specialist audience reading a technical guide. A grade 5 score can read as patronising in the wrong place.
• Search performance. Readability scores are not a ranking factor, and no target score will raise a page's position or traffic. Write clearly because readers benefit, not to hit a number.
• Languages other than English. The weights were fitted to English, and the syllable rule is English-only.

A practical way to use them

1. Paste a whole section, not a single sentence, into the calculator.
2. Look at words per sentence and syllables per word before the scores. They tell you which kind of edit will help.
3. If words per sentence is high, find the longest sentences and split the ones that carry two ideas.
4. If syllables per word is high, look for padding phrases and abstract nouns ("facilitate the optimisation of") and replace them with the verb you mean.
5. Re-check the sentence count if the text has abbreviations or decimals.
6. Stop when the paragraph reads well aloud, whatever the score.

The same calculator shows reading time for the text you paste, explained in how reading time is calculated. For headlines, where a single line is too short for these formulas to mean much, the Headline Analyzer looks at length and word choice instead, with its own caveats.