You paste your text into a readability tool and it tells you it’s written at a “Grade 11” level. Is that good? Bad? Should you aim for Grade 5? Grade 8? And what does “grade level” even mean when the reader is a 40-year-old professional?
Readability formulas are useful, but they’re frequently misapplied because people treat the output as a quality score rather than what it is: a rough proxy for sentence and word complexity. Here’s how these formulas actually work and how to use them without overthinking it.
How Readability Formulas Work
Most readability formulas measure two things: how long your sentences are, and how many syllables (or characters) your words have. That’s it. They don’t measure clarity of ideas, logical structure, accuracy, tone, or whether your writing actually communicates well. They measure surface-level complexity.
Flesch-Kincaid Grade Level
The most commonly cited formula. It was developed in 1975 by J. Peter Kincaid and his team for the US Navy, which needed a way to assess whether technical manuals were readable by enlisted sailors with varying education levels.
The formula: 0.39 × (total words / total sentences) + 11.8 × (total syllables / total words) - 15.59
The output is a US school grade level. “Grade 8” means the text should be understandable by an average eighth-grader. The formula rewards shorter sentences and shorter words, which is a reasonable heuristic for accessibility but nothing more.
Flesch Reading Ease
The inverse of Kincaid: higher scores mean easier reading. The scale runs from 0 to 100:
- 90-100: Very easy. Short sentences, simple words. Think children’s books.
- 60-70: Standard. Comfortable for most adults. Most journalism falls here.
- 30-50: Difficult. Long sentences, multi-syllable vocabulary. Academic papers.
- 0-30: Very difficult. Essentially requires specialized knowledge to parse.
Gunning Fog Index
Created by Robert Gunning in 1952. Similar to Flesch-Kincaid but defines “complex words” as those with three or more syllables (excluding common suffixes like -ing, -ed, -es). The output is also a grade level.
Coleman-Liau Index
Uses character count per word instead of syllable count, which makes it easier to compute programmatically (counting characters is simpler than counting syllables). The output is a grade level.
All of these formulas are measuring the same underlying thing with slightly different math. If one says your text is Grade 9 and another says Grade 11, that’s normal. The absolute number matters less than the ballpark.
Why Web Writing Should Target a Lower Grade Level
The most common advice is to write web content at a Grade 6-8 level. This sounds like you’re writing for children, and many writers resist it. But the recommendation isn’t about the reader’s intelligence. It’s about the reading context.
People don’t read web content the way they read a book. They scan. They’re often on a phone, distracted, or reading in a second language. Research from the Nielsen Norman Group consistently shows that web users read about 20-28% of the text on a page. They’re making rapid decisions about whether to keep reading, click away, or scroll past.
Short sentences and common words reduce the cognitive effort per scan. A reader who is skimming can still extract the key points from a Grade 7 passage. The same information buried in a Grade 12 passage might be missed entirely, not because the reader couldn’t understand it if they stopped and read carefully, but because they won’t stop and read carefully. That’s the reality of web reading behavior, and fighting it is a losing strategy.
For comparison, here are the approximate reading levels of some well-known publications:
- USA Today: Grade 6-7
- New York Times: Grade 8-10
- Harvard Business Review: Grade 11-13
- Academic journals: Grade 14-18
The NYT and HBR write for educated audiences but still keep things below Grade 13. Their readers could handle denser prose, but denser prose would reduce engagement and comprehension, so the publications don’t write that way.
Where the Formulas Fail
Readability formulas are crude tools. They’re useful directionally but misleading if you take the numbers too literally.
They Don’t Measure Clarity
The sentence “The implementation of the organizational restructuring initiative is pending” scores worse than “We’re changing how the team is set up,” but not because of sentence length or word complexity alone. It’s the abstraction and passive voice that make it harder to understand. Readability formulas catch some of this (longer words, more syllables) but miss the deeper problem.
Conversely, you can write a confusing, incoherent paragraph using only short words and short sentences. The formula would give it a good score.
They Penalize Technical Vocabulary Unfairly
If you’re writing about “photosynthesis” or “amortization” or “mitochondria,” you need those words. They’re polysyllabic, which hurts your readability score, but replacing them with simpler words would make the writing less precise and less useful to the reader who searched for that topic.
The fix is to use technical terms where necessary but keep the sentence structure around them simple. “Amortization is how your loan payments are split between interest and principal” uses one complex word in a simple sentence. The readability score won’t be perfect, but the text is clear.
They Reward Choppy Writing
Taken to an extreme, optimizing for readability scores produces writing that sounds robotic. Short sentence. Then another. Simple words only. No subordinate clauses. This reads terribly despite scoring well.
Good writing has rhythm: a mix of sentence lengths and structures. A long sentence that flows well is often easier to read than three short sentences that interrupt the reader’s rhythm.
How to Actually Use Readability Scores
Use them as a check, not a target. Write naturally, then check the score. If it’s above Grade 10 for web content, look for sentences you can split and words you can simplify. Don’t rewrite until the number hits a specific target.
Compare drafts to each other. If your revision scores two grade levels lower than your first draft, you probably made it more accessible. The relative change is more informative than the absolute number.
Look at sentence length separately. Average sentence length above 25 words is a reliable signal that some sentences should be broken up. This is more actionable than the composite score.
Ignore the score for fiction, poetry, and creative writing. These formulas were designed for expository and technical writing. Applying them to creative work is meaningless.
Don’t chase perfection. A Grade 8 article that clearly explains a complex topic is better than a Grade 5 article that oversimplifies it. Clarity and accuracy trump the number.
Check Your Own Writing
Try the Readability Analyzer on ToolzHQ to see where your text lands, or use the Word Counter to check sentence length and structure before you publish.