Six classical formulas are calculated from three measurements: how many words there are, how many sentences they are spread across, and how long the words themselves are. Flesch Reading Ease is reported first as a score out of a hundred, where higher is easier. The other five report a US school grade, which is a different scale for the same underlying idea.
Two of the six — Coleman-Liau and the Automated Readability Index — count letters rather than syllables. They are included precisely because English syllable counting has no correct answer, and seeing a letter-based score beside the syllable-based ones shows how much of the result depends on that guess. Below the scores, the measurements that produced them are listed along with the longest sentence, which is almost always the single most useful thing on the page.
What these formulas do and do not measure
-
They measure sentence length and word length:
not clarity, not accuracy, and not whether the writing is any good. A page of short sentences saying nothing scores well, and a precise technical explanation scores badly for being precise.
-
Syllable counting is a heuristic:
fire, every and business are counted differently by different tools, so scores vary by a point or two between them. Anyone reporting a readability score to two decimal places is overstating what it knows.
-
English only:
the formulas were built between the 1940s and 1970s on American schoolbooks and military manuals. Applying them to Urdu, German or any other language produces a number with no meaning behind it.
-
Treating the score as a target produces worse writing:
chopping sentences to move a number is the same mistake as padding an article to hit a word count. Use the score to find the sentence that needs splitting, not to hit a figure.
Six formulas, three measurements, and one guess buried inside four of them
Every score on the page comes from the same three counts: how many words, how many sentences, and how long the words are. Flesch Reading Ease turns those into a score out of a hundred where higher means easier; the [msm_tool slug="readability-scorer"] then reports five more numbers — Flesch-Kincaid Grade, Gunning Fog, SMOG, Coleman-Liau and the Automated Readability Index — each expressed as a US school grade, which is the same underlying idea on a different scale.
Four of those six measure word length by counting syllables, and English syllable counting has no single correct answer: fire, every and business are split differently by different counters, which is why two readability tools can disagree by a point or two on identical text without either being wrong. Coleman-Liau and the Automated Readability Index sidestep the argument entirely by counting letters instead. They are included specifically so a letter-based score sits next to the syllable-based ones — the gap between them is a rough measure of how much of the result is guesswork.
The number that matters more than any score
Each score is reported alongside the measurement that produced it, and the longest sentence in the text. That sentence is usually the single most useful thing on the page, because splitting it is the change most likely to move every other number at once.
What a good score does and does not mean
These formulas measure sentence length and word length. They do not measure clarity, accuracy, or whether the writing says anything. Short sentences that say nothing score well; a precise technical explanation scores badly for being precise. Treating the number as a target produces the same failure as padding an article to hit a word count — chopping sentences to move a score is not the same thing as improving the writing.
Built on English, and only English
All six formulas were built between the 1940s and 1970s on American schoolbooks and military training manuals. Running Urdu or German text through them still produces a number, but not one that means anything — the formulas assume a language they were never built to measure.
Where SMOG needs a caveat
SMOG was designed around a sample of thirty sentences, and its constant assumes that sample size. Under thirty sentences it still produces a figure, and that figure should be read as indicative rather than as a measurement — the shorter the passage, the less it means.
A working target, not a rule
As a rough guide, a Flesch score around sixty or higher suits a general audience, and seventy or higher suits marketing copy; technical documentation legitimately sits lower than either. None of that is a rule any more than a word count is one — it is a starting point for deciding whether a specific sentence needs to change, not a figure to write toward.