What a Flesch Reading-Ease Score Actually Tells You
A Flesch score is just arithmetic on two things: how long your sentences are and how many syllables your words have. That's it. No comprehension testing, no semantic analysis, no sense of whether your argument makes sense. Understanding that limitation is the key to using the score well instead of chasing it blindly.
The formula, in plain terms
The Flesch Reading-Ease formula takes your average sentence length (words per sentence) and your average word length (syllables per word), and combines them into a single number, typically between 0 and 100. Higher numbers mean the text is generally easier to read; lower numbers mean it's denser. The exact formula involves specific multipliers, but you don't need to memorize them — the concept is what matters: short sentences and short words push the score up, long sentences and long, syllable-heavy words push it down.
That's the entire mechanism. It doesn't know if your sentence is well-argued, factually correct, or even grammatically sound. It's purely counting.
Why that narrow focus is still useful
Sentence length and word length correlate reasonably well with how much working memory a sentence demands from a reader. A 40-word sentence stuffed with multisyllabic words asks a reader to hold a lot in their head before the meaning resolves. A 12-word sentence with common words resolves almost immediately. That correlation is real, which is why the score, despite measuring something narrow, tends to track something readers actually feel.
The mistake is treating that correlation as identical to quality. Plenty of excellent writing scores lower than the average blog post because the subject matter demands precision that shorter, simpler words can't deliver. Plenty of bad writing scores high because short, choppy sentences with simple words can still fail to make a coherent point.
Reading the score bands
Most tools that report a Flesch Reading-Ease score use roughly the same band interpretations, adapted from the original research into what reading level each band corresponds to.
| Score range | Typical audience / reading level |
|---|---|
| 90-100 | Very easy; understandable by an average 11-year-old |
| 70-89 | Easy to fairly easy; comfortable for most general readers, roughly middle-school to early high-school level |
| 60-69 | Plain English; the common target for general web content and news writing |
| 50-59 | Fairly difficult; comfortable for high-school graduates, more effort for casual browsing |
| 30-49 | Difficult; college-level reading, common in more technical or academic content |
| 0-29 | Very difficult; typically graduate-level or highly specialized/technical text |
Where the score is genuinely useful
For most general web writing — blog posts, marketing copy, how-to guides — landing somewhere in the 60-70 range is a reasonable target because it signals the writing isn't creating unnecessary friction. If your score is consistently in the 20s and 30s for content meant for a general audience, that's a real signal worth acting on: your sentences are probably longer and more clause-heavy than they need to be for the job the content is doing.
The score is also genuinely useful as a consistency check across a site or a piece — if one section of an article scores dramatically lower than the rest, that's often a sign it needs a second pass, even if you can't immediately articulate why it feels harder to read.
Where the score misleads you
If you're writing for a technical, academic, or specialist audience, a lower score isn't a failure — it might be accurate to the subject. Legal writing, medical documentation, and academic papers routinely score in the 20s and 30s, not because they're badly written but because precision in those fields sometimes requires longer sentences and more specific vocabulary than a general-audience score expects.
The other trap is chasing the number directly. If you take a well-constructed paragraph and mechanically chop every sentence in half to raise the score, you often get choppier, more disjointed writing that reads worse despite scoring "better." The score should follow from clear writing, not drive it.
Reading Ease versus Grade Level
Flesch Reading Ease has a close relative called the Flesch-Kincaid Grade Level, and the two get confused constantly because they're built from the same underlying inputs — average sentence length and average syllables per word. The difference is purely in how the result is presented. Reading Ease outputs a 0-100 score where higher means easier. Grade Level translates the same underlying math into an estimated U.S. school grade, where a score of 8 roughly corresponds to eighth-grade reading level. Some tools report one, some report both, and it's worth knowing they're two views of the same calculation rather than two independent measurements, so there's no need to reconcile a discrepancy between them as if it were a bug — they should track each other consistently precisely because they share the same formula inputs.
Why syllable count is doing more work than it seems
It's tempting to assume word length matters less than sentence length, but syllable count is often the more decisive factor in practice. Two sentences with identical word counts can score very differently if one uses short, common words ("get," "use," "help") and the other leans on longer, more formal vocabulary ("obtain," "utilize," "facilitate"). This is part of why the score sometimes surprises writers who feel like they've written a fairly short, punchy sentence, only to see it register as denser than expected — the sentence might be short in word count but heavy in syllables, and the formula weighs both.
This is also a useful editing lever independent of sentence length. If a piece is scoring lower than you'd like and restructuring sentences feels excessive, scanning for unnecessarily long words with shorter, equally accurate alternatives is often a faster fix — swapping "utilize" for "use," or "facilitate" for "help," moves the score without touching sentence structure at all.
How to actually use it
Treat the Flesch score as a diagnostic, not a target. Write the piece for your actual audience first. Then check the score, and if it's wildly out of step with what that audience needs — too dense for a general reader, too simplified for a technical one — go back and look at specific sentences rather than trying to hit a number in the abstract. The goal was never the number. The number is just a fast way to notice when something might be off.
You can run any draft through Grammachine's readability checker to get a Flesch score alongside other readability signals, which is a faster way to catch outlier sentences than reading a whole piece aloud looking for dense patches.
For more breakdowns on readability and editing craft, sign up at /#lead.
Frequently asked questions
What is a 'good' Flesch Reading-Ease score?
It depends entirely on your audience and purpose. General web content usually aims for 60-70, but technical or academic writing legitimately scores lower without being poorly written.
Does the Flesch score measure how good my writing is?
No. It measures a narrow thing — sentence length and word length — as a proxy for reading difficulty. It says nothing about clarity of argument, tone, accuracy, or whether the content is actually useful.
Can I game the score by just shortening sentences?
You can move the number, but chopping sentences arbitrarily to hit a target often makes writing choppier and harder to follow, which defeats the purpose of chasing readability in the first place.
Is a higher score always better?
Not necessarily. A very high score (90+) can sometimes indicate writing that's been oversimplified to the point of losing nuance, depending on the subject matter.
What's the difference between Flesch Reading Ease and Flesch-Kincaid Grade Level?
They use the same underlying inputs (sentence length, syllables per word) but present the result differently — Reading Ease gives a 0-100 score where higher is easier, while Grade Level translates the same math into a U.S. school grade estimate.