A score you can fake in ten seconds
Here’s a passage: “Cat sat. Dog ran. Sun rose. Bees buzzed. Wind blew. Cars honked. Kids yelled. Doors slammed.”
Run it through a Flesch Reading Ease calculation and it scores 120.2, well above the “very easy” ceiling of 100. The grade level comes out below zero. The Gunning Fog Index sits well under one.
By the numbers alone, this is the easiest, clearest writing you could possibly produce. Read it out loud, though, and you’ll notice it isn’t writing at all. It’s eight disconnected fragments with no subject, no thread, and no reason to sit next to each other. A formula built entirely from word length and sentence length can’t tell the difference between plain, connected prose and a list of unrelated nouns and verbs. Short sentences score well. That’s true whether or not those sentences add up to anything.
This isn’t a bug in one particular formula. It’s true of Flesch Reading Ease, Flesch-Kincaid Grade Level, and Gunning Fog alike. All three are built from the same two ingredients: how long the words are, and how long the sentences are.
Readability formulas fail in both directions
Gaming a score toward “easy” is one direction. The other direction fails just as badly, and it’s more common in real writing.
A sentence using precise field-specific terms can score poorly, even when it’s exactly the right way to say something to its intended reader. A cardiology report or a tax statute reaches for long, specific words because plain synonyms don’t exist, or because precision matters more than accessibility. Scoring that sentence as “difficult” isn’t wrong on the formula’s own terms. Treating the score as a verdict on the writing itself is wrong, because it misses context a formula was never built to see.
This is why grade-level targets need judgment, not blind application. A children’s book and a surgical consent form both deserve careful writing. They don’t deserve the same target score, and no formula can tell you which one you’re looking at.
What a readability score cannot see
A readability score is a measurement of surface features. It counts letters, syllables, and punctuation. It never reads for meaning, which means entire categories of difficulty are invisible to it.
Word familiarity. “Cat” and “cetacean” are both one syllable, but only one of them is instantly recognizable. A score treats them identically.
Sentence structure. A short sentence with a buried subject can be harder to parse than a longer, plainly built one. Formulas count words, not grammar, so a sentence with an unusual structure gets no penalty for being confusing.
Coherence between sentences. Writing isn’t just a string of individual sentences. It’s a sequence where each one should build on the last. A formula scores each sentence’s word count in isolation. It has no way to notice when three connected sentences would communicate more clearly than five choppy ones.
Actor clarity. Passive voice, vague pronouns, and buried subjects can all make a sentence harder to follow, without adding a single extra syllable or word. None of that shows up in a formula built purely from counts.
Why one number was never going to be enough
None of this makes readability formulas worthless. They’re a fast, honest signal for one thing: whether your words and sentences run long or short on average. That’s genuinely useful information, and it takes seconds to compute.
The mistake is treating that one number as a full verdict on writing quality. A score search that stops at “sounds hard” or “sounds easy” misses the more useful question: which specific sentence is causing the problem, and why? Two documents can share an identical Flesch score and read completely differently, because the score never looked at what either one actually says.
Why this site pairs scores with sentence-level detail
That gap is the reason our readability explainer doesn’t stop at a score. It highlights the sentences actually dragging your text down. Then it explains the specific issue in each one. Maybe it’s a passive construction with no clear actor. Maybe it’s a nominalization hiding a verb, or a sentence where the subject and verb sit too far apart to track easily.
A score tells you there might be a problem. Sentence-level detail tells you exactly where it is, and gives you a concrete way to fix it. If your writing scores lower than you’d like, the fix usually isn’t “use shorter words.” It’s finding the two or three sentences actually causing the trouble, which a single number can never point to on its own.
Read our companion piece on why long sentences are hard to follow for a closer look at one of the most common culprits. Or run your own text through the readability explainer to see the detail behind your own score.