Calculate all 6 readability scores at once
Free tool runs Flesch Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau, and ARI on your text instantly.
What Are Readability Scores?
Readability scores are numerical estimates of how easy a piece of text is to read, expressed either as a US grade level (e.g. “8.2” means a mid-8th grader can read it) or as a 0 to 100 ease score where higher means easier. They are calculated from surface features of the text: sentence length, word length, and syllable or character counts.
Writers, marketers, educators, government communicators, and healthcare professionals use them to make sure their words reach the audience they were written for. The US Department of Defense requires Flesch-Kincaid Grade Level on technical manuals. The CDC and NIH publish patient-education guidelines anchored in readability scores. Insurance contracts in many US states must legally meet a minimum readability threshold.
The formulas matter because writing that scores too high excludes readers, while writing that scores too low can feel condescending. A good readability tool gives you both signals at once.
Flesch Reading Ease (1948)
Rudolf Flesch's original 1948 formula. It outputs a score from 0 to 100, where higher means easier. It is the only widely-used readability score that is not a grade level, and it remains the most cited readability metric in writing tools.
Formula
206.835 − 1.015 × (words / sentences) − 84.6 × (syllables / words)| Score | Difficulty |
|---|---|
| 90-100 | Very easy |
| 80-89 | Easy |
| 70-79 | Fairly easy |
| 60-69 | Standard |
| 50-59 | Fairly difficult |
| 30-49 | Difficult |
| 0-29 | Very difficult |
For most adult content (blog posts, product copy, internal docs), aim for 60 to 70. Newspapers like USA Today sit around 65. Reader's Digest is around 75.
Flesch-Kincaid Grade Level (1975)
Commissioned by the US Navy in 1975 from J. Peter Kincaid and colleagues. It uses the same inputs as Flesch Reading Ease (sentence length and syllables per word) but rescales the output as a US grade level. A score of 8.2 means roughly a mid-8th grader can read it comfortably.
Formula
0.39 × (words / sentences) + 11.8 × (syllables / words) − 15.59Flesch-Kincaid is the formula required by US Department of Defense documentation standards (MIL-STD-1689) and by many state insurance regulators. It is also the default grade-level score in Microsoft Word, Google Docs, and Grammarly, which is why most writers encounter this number first.
Gunning Fog Index (1952)
Robert Gunning developed the Fog Index in 1952 while consulting with newspaper and business publications. He used “fog” as a metaphor for confusing, jargon-heavy prose. The output is the years of formal education a reader needs to understand the text on first reading.
Formula
0.4 × ((words / sentences) + 100 × (complex_words / words))“Complex words” are words of 3 or more syllables, with three exclusions:
- Proper nouns (“Anthropic”, “California”) do not count
- Compound words made of simple parts (“bookkeeper”) do not count
- Common verb endings (-es, -ed, -ing) that push a word to 3 syllables do not count
Gunning Fog is popular in business writing because it specifically penalizes jargon. A target of 7 to 8 means most readers can finish a sentence on the first read. Above 12 and you are losing audience.
SMOG (1969): Simple Measure of Gobbledygook
Harry McLaughlin's 1969 formula. SMOG stands for Simple Measure of Gobbledygook, named in deliberate homage to Gunning's “fog.” It is the formula most commonly used in healthcare communication research because McLaughlin calibrated it specifically against 100% comprehension (versus the 50-75% that Flesch-Kincaid implies).
Formula
1.0430 × √(polysyllables × 30 / sentences) + 3.1291“Polysyllables” here means words of 3 or more syllables, with no exclusions (unlike Gunning Fog). The output is a US grade level.
Important caveat
SMOG was designed for samples of at least 30 sentences (specifically: three 10-sentence chunks from the beginning, middle, and end of a document). On shorter text, the polysyllable density swings too much and the resulting grade level is unreliable. For paragraphs or short emails, prefer Flesch-Kincaid, Coleman-Liau, or ARI.
Coleman-Liau Index (1975)
Meri Coleman and T. L. Liau's 1975 formula. Its key innovation: it uses character counts instead of syllables, eliminating the most error-prone step in older formulas. This makes it ideal for text where syllable counting is unreliable: technical writing, abbreviations, code samples, scientific notation, and any text with unusual orthography.
Formula
0.0588 × L − 0.296 × S − 15.8L = average letters per 100 words. S = average sentences per 100 words. Output is a US grade level.
Because syllable counts in software depend on imperfect dictionaries or estimation algorithms, Coleman-Liau tends to be more consistent across automated tools. If two readability checkers disagree on Flesch-Kincaid but agree on Coleman-Liau, the Coleman-Liau number is usually the more trustworthy one.
Automated Readability Index (ARI, 1967)
Edgar Smith and R. J. Senter built ARI in 1967 for the US Air Force. Its design goal was speed: ARI was meant to be calculated in real time on the electric typewriters and early computer terminals of the late 1960s. It uses character and word counts only, no syllables.
Formula
4.71 × (chars / words) + 0.5 × (words / sentences) − 21.43ARI and Coleman-Liau measure essentially the same thing through slightly different math. They usually agree within 1 grade level. Use ARI when speed matters or when you want a second character-based opinion alongside Coleman-Liau. The two formulas together are a good sanity check on the syllable-based scores.
Which Score Should You Trust?
Best practice: average all five grade-level formulas (Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau, ARI). Skip Flesch Reading Ease in the average because it uses a different scale. The mean of the five gives the most robust estimate, because each formula has known biases:
Flesch family
Overweights syllables. Penalizes Latin and Greek-derived technical vocabulary even when readers know the terms.
Gunning Fog
Flags compound business jargon as “complex” even when fluent readers handle it without effort.
SMOG
Skews high (more conservative). Designed for 100% comprehension, so it predicts a higher grade than other formulas.
Coleman-Liau / ARI
Skip syllables. Excellent for code, abbreviations, and technical text. Slightly less accurate on standard prose.
When the five grade-level scores disagree by more than 2 levels, the text has unusual properties (heavy jargon, lots of abbreviations, very short sentences). Read the passage yourself. The formulas are useful, but they are not a substitute for editorial judgment.
Recommended Reading Levels by Audience
Targets vary by audience and purpose. The grade-level column below is the average of the five grade-level formulas. The Flesch Ease column is the corresponding 0-100 score.
| Audience | Grade level |
|---|---|
| Children's content | 3rd-5th |
| General public / mass media | 6th-8th |
| Marketing / SEO blog content | 7th-9th |
| News (NYT, WaPo) | 9th-12th |
| Medical patient education (CDC/NIH) | 6th-8th |
| Technical / engineering | 12th-16th |
| Academic papers | 13+ |
| Legal documents | 13th-17th |
Note that medical patient education has the same target as mass media (6th-8th grade), even though the underlying content is more complex. That is the entire point of the CDC and NIH guidelines: clinical accuracy delivered at a grade level any patient can act on.
How to Improve Readability
Shorten sentences
Target an average of 15 to 20 words per sentence. Sentence length is the single biggest factor in every readability formula. Break compound sentences with periods, not commas.
Use simpler words
Prefer 1-2 syllable words when they exist. “Use” beats “utilize.” “Help” beats “facilitate.” “Show” beats “demonstrate.” This single change moves Flesch-Kincaid down by 1-2 grade levels.
Active voice over passive
“The team shipped the feature” reads faster than “The feature was shipped by the team.” Active voice cuts word count and clarifies who did what.
Cut adverbs and qualifiers
Words like “very,” “really,” “quite,” and “basically” add length without adding meaning. Strong verbs do the work better.
Bullet points and lists
Lists score very well on every formula because the parser treats each item as a short sentence. They also genuinely improve scanning, which formulas approximate.
Headers every 2-3 paragraphs
Subheadings break long passages, give skimmers waypoints, and (because each is its own short “sentence”) drop the average sentence length used by every formula.
Read aloud test
If you cannot read a sentence aloud without pausing for breath or rereading, simplify it. This is the oldest readability test and still the most reliable.
Calculate all 6 readability scores for your text instantly with our Readability Checker.
Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau, and ARI side by side, with audience targets and a single averaged grade level.
Open Readability CheckerFrequently Asked Questions
For general adult content, target a Flesch Reading Ease score of 60 to 70 (the “standard” band, roughly 8th to 9th grade). For mass-market or healthcare content, push above 70. Academic and legal text often scores 30 to 50, which is acceptable for that audience but unreadable for the public.
For mass audiences, target 6th to 8th grade. The CDC and NIH recommend 6th to 8th grade for patient education. News outlets like The New York Times typically score 9th to 12th. Marketing and SEO blog content sits comfortably at 7th to 9th grade. Anything above 12th excludes most general readers.
Each formula weights inputs differently. Flesch family formulas count syllables, which penalizes Latin and Greek-derived technical vocabulary. Gunning Fog flags any 3+ syllable word as complex. SMOG skews higher and is designed for long samples. Coleman-Liau and ARI use character counts and skip syllable estimation entirely. The same paragraph can vary by 2 to 4 grade levels across formulas.
Flesch-Kincaid is reasonably accurate for English prose and is the formula required by US military and federal documentation standards. It correlates well with reader comprehension on standard text. It is less reliable for technical jargon, code samples, lists, and very short passages. For best results, run it on samples of at least 100 words.
SMOG was designed by Harry McLaughlin to be calculated on three 10-sentence chunks (one from the beginning, middle, and end of a document). With fewer than 30 sentences, the polysyllable count varies too much from sample to sample and the resulting grade level becomes unstable. On short text, prefer Flesch-Kincaid, Coleman-Liau, or ARI.
Yes, dramatically. Most readability formulas were calibrated on English. They give misleading results for German (long compound nouns), Spanish (different syllable distribution), or Chinese (no syllables in the same sense). Language-specific formulas exist (LIX for Swedish, Fernandez Huerta for Spanish, Kandel-Moles for French) and should be used instead.
Yes. Scoring 3rd grade for an adult professional audience reads as condescending or low-effort. Match the score to the audience: children's content benefits from very low scores, but a SaaS landing page targeting CFOs at 4th grade feels off-tone. Aim for the lowest score that still respects the reader's expertise.
They are estimates, not measurements. Formulas count surface features (sentence length, syllables, characters) and ignore meaning, structure, vocabulary familiarity, and reader background knowledge. A passage about quantum mechanics with short sentences can score 6th grade and still be incomprehensible. Use scores as one signal among several, not a verdict.