Open any writing tool and you will see a bewildering shelf of readability scores: Flesch Reading Ease, Flesch–Kincaid Grade Level, Gunning Fog, SMOG. They sound like competing verdicts. In practice, they are four cousins measuring the same two things from slightly different angles. This guide explains each score in plain English, why they sometimes disagree, and which one to watch for your kind of writing.
The shared intuition behind all four
Every readability formula is built from the same two ingredients: sentence length and word difficulty. A sentence with more words asks the reader to hold more. A word with more syllables is more likely to be rare and hard. The formulas weigh these ingredients differently, which is why they rarely agree exactly — and why none of them is “wrong”.
The four scores, decoded
Flesch Reading Ease (0–100)
Created by Rudolf Flesch in 1948, this is the oldest and most famous score. It counts words per sentence and syllables per word, then maps them onto a 0–100 scale: the higher the number, the easier the text. Roughly, 90–100 is elementary, 60–70 is plain English (the level of Reader's Digest and most news), and 30 or below is dense academic prose.
Flesch–Kincaid Grade Level
Introduced for US Navy training materials, this converts the same measurements into a US school grade. A score of 8 means an average eighth-grader can read it. It is the score most commonly quoted in plain-language guidelines because “grade 8” is a concrete, memorable target.
Gunning Fog Index
Robert Gunning's 1952 formula adds a twist: it treats every 3+ syllable word as a “hard word” and estimates the education needed to read the passage. Because it counts hard words so bluntly, jargon-heavy prose scores much higher on Fog than on Flesch–Kincaid.
SMOG Grade
Developed by G. Harry McLaughlin in 1969, SMOG samples every tenth sentence and counts polysyllables even more aggressively. It is famously strict — usually 1–2 grades higher than Flesch–Kincaid — which makes it popular in health literacy work, where underestimating difficulty is dangerous.
Why the scores disagree
Take a sentence full of short words but structured in a long chain: “We met, reviewed the options, argued, and finally chose a path.” Flesch–Kincaid sees long sentences and scores it moderately; SMOG counts almost no polysyllables and scores it easy. Conversely, a short sentence stuffed with jargon — “The cortical reuptake attenuated.” — trips every hard-word counter at once. Disagreement is information: which ingredient is hurting you tells you what to fix.
Target ranges by format
| Format | Flesch target | Grade target |
|---|---|---|
| Blog posts, emails, landing pages | 60–80 | Grade 6–8 |
| News articles, business reports | 50–70 | Grade 8–10 |
| Academic papers, legal documents | 30–50 | Grade 12–16 (expected) |
| Instructions, product help | 70–90 | Grade 5–7 |
Notice what the table does not say: academic writing should not target grade 8. Density is sometimes the correct genre choice. Readability targets exist to keep your difficulty in line with your audience, not to flatten every genre into plain English.
How StyleScope computes them
The Readability Score tool and the full analyzer count your words, sentences and syllables with a fast rule-based counter, then apply the standard published formulas — the same ones used by word processors worldwide. The syllable counter is an approximation (English pronunciation is irregular), so treat scores as reliable estimates within about one grade point, not laboratory measurements.
Using the scores as a routine
- Draft without looking at the numbers. Write the thinking first.
- Run the analyzer once. Note Flesch and grade, and the gap between the two grade scores.
- Fix the dominant problem. High Fog but okay Flesch → swap jargon. Both high → split sentences.
- Re-run and confirm. Watch the grade drop one point at a time; chasing three grades at once usually means you have rewritten the voice, not just the difficulty.
A short history (why these formulas persist)
It is worth asking why seventy-year-old formulas still power modern writing tools. The answer is that they measure something stable: human short-term memory does not stretch, and rare words stay rare. Every newer “AI-grade” readability model that claims to do better ultimately predicts reader effort from the same two signals plus a handful of refinements — pronoun density, passive constructions, sentence-connection types. If you want one number to watch while drafting, the classic formulas remain the most honest trade-off between simplicity and usefulness.
What scores cannot tell you
Scores know nothing about logic, evidence, beauty or truth. A perfectly readable sentence can be wrong; a dense paragraph can be the best thing you ever write. Readability metrics are a floor, not a ceiling — they make sure your reader is not spending effort on decoding when they should be spending it on understanding. Keep them in that role, and they are quietly one of the most reliable tools a writer owns.