This guide aims to show how different readability formulas work, which tools calculate them for free, and why the score matters for both SEO and accessibility.
Words are important to me. I’m a former ESL teacher and a reasonably experienced long-form content writer (if I do say so myself, which I did, just then). Language has power, and everyone deserves to have access to it.
Writing for high readability means shortening sentences and swapping complex words until a formula such as Flesch-Kincaid or SMOG returns a target score, usually an eighth-grade reading level or lower for public-facing content. Run the draft through a scoring tool, cut sentences over 20 words, and re-test until the number holds.
Readability formulas
Every readability formula determines its score by running some math on word or syllable difficulty, average sentence length, and, for some formulas, a check against a list of familiar words. Feed those numbers into the formulas and out comes one of two outputs: a 0-100 reading-ease score where higher means easier, or a grade-level number that maps to a U.S. school year (like an 8.2 for text a typical eighth-grader could follow).
A formula can’t tell you if the writing is “good” or flows well. It only predicts how hard the words and sentences are to process. The idea traces back to Edward Thorndike’s 1921 word-frequency research, which showed that readers recognize some words faster than others regardless of meaning.
I also want to make it clear that this type of “readability” is not “legibility” and assumes the reader can physically read the words. Things like font size, line height, and letter spacing are ignored by these text-based formulas. A paragraph can score well on these tests and still be impossible or very difficult for someone to read if that paragraph has insufficient text color contrast.
The five major formulas
These five formulas dominate practical readability work. Here’s how they all work:
Flesch Reading Ease
206.835 minus 1.015 times average sentence length, minus 84.6 times average syllables per word, on a 0-100 scale where higher means easier.
Flesch-Kincaid Grade Level
Runs the same two inputs (average sentence length and average syllables per word) through 0.39 times sentence length plus 11.8 times syllables per word, minus 15.59. It reports a U.S. school grade instead of a 0-100 score.
Gunning Fog
Adds word complexity measured as a percentage (not just syllable count). The formula multiplies 0.4 by the sum of average sentence length and the percentage of words with three or more syllables.
SMOG
Multiplies 1.0430 by the square root of 30 times polysyllabic words per sentence, then adds 3.1291.
Dale-Chall
Checks each word against a list of roughly 3,000 words a fourth-grader knows, then combines the percentage not present in that list with average sentence length. For example: A word like “onus” has only two syllables and would barely move a syllable-based score, but it isn’t on Dale-Chall’s familiar-word list, so it drags that score up regardless.
Grade-level scores for familiar publications work as calibration points. Reader’s Digest and mainstream news/magazine writing tends to sit toward the accessible end of Flesch’s original scale, around an 8th-grade reading level.
Here’s the scores for the opening paragraph from Moby Dick so you can see how these formulas all score differently:
| Formula | Score | Meaning |
| Flesch Reading Ease | 54.5 | Fairly difficult, college level |
| Flesch-Kincaid Grade Level | 11.9 | 11th-12th grade |
| Gunning Fog | 14.43 | College sophomore |
| SMOG | 12.6 | Upper high school (extrapolated from the short sample) |
| Dale-Chall | 9.5 | 9th-10th grade |
As you can see, five formulas built from the same two or three surface variables can produce wildly different grade levels from a single paragraph.
None outranks the others. Each weighs sentence length and word difficulty differently, so editorial teams that track the number usually check more than one before trusting it.
WCAG and readability for accessibility
WCAG 2.2 Success Criterion 3.1.5, Reading Level, sets a threshold.
Text shouldn’t demand reading ability “more advanced than the lower secondary education level” unless the page also offers a simplified version. It sits at Level AAA, the strictest conformance tier, which is why most audits skip it even when a site claims full WCAG 2.2 AA. The criterion names no formula, but auditors typically reach for Flesch-Kincaid or SMOG as the working stand-in for “lower secondary.”
The Plain Writing Act of 2010 approaches the same problem from government’s side. It requires federal agencies to draft public-facing documents in language readers can understand on a first pass. Agencies must also report on how well they’re doing it. The Department of Veterans Affairs applies that mandate to patient-facing materials, checking discharge instructions and consent forms against readability scores. A misread dosage line is a real risk, not a style complaint. Government contractors and public agencies that fail SC 3.1.5 carry the same exposure as any other WCAG failure. That means complaints, demand letters, and in serious cases an OCR letter, not just a note about awkward prose. W3C’s draft WCAG 3.0 is rewriting readability testing from the ground up, so confirm which version of the standard is live before building a compliance program around 3.1.5.
Readable text ranks better
Text scored at a lower grade level keeps more readers on the page, because fewer of them quit mid-paragraph. That shows up in dwell time and bounce rate, engagement signals Google’s ranking systems have tracked for years. Featured snippets and voice-search answers compound the effect: both formats tend to pull short, syntactically simple sentences, which favors writing in roughly the 6th- to 8th-grade range. A sentence loaded with subordinate clauses rarely gets lifted into a snippet box, however accurate it is.
Google’s own guidance on writing content for search engines tells writers to produce content for people first, answering a reader’s question rather than chasing a keyword count.
Readability isn’t a ranking factor Google’s algorithms check directly, but it functions as a proxy for the engagement signals search engines do measure. The same clear, direct sentences that score well on readability formulas also tend to be the ones readers finish instead of abandoning.
Content readability checklist
Knowing a paragraph’s readability score is one thing. Getting it down to a target grade level without stripping the meaning out is another, and it takes practice to get it right. Four steps to work on it yourself are:
- Cut any sentence over 20 words, or split it with a comma or conjunction.
- Swap three-plus-syllable words for shorter synonyms when possible (“utilize” to “use,” “facilitate” to “help”).
- Break paragraphs longer than four sentences into two paragraphs.
- Add a subhead roughly every 300 words.
The target grade level depends on the audience, not a universal minimum. General web copy reads best around 7th to 8th grade. Consumer health information and government-facing content should drop to 5th or 6th grade. Technical documentation and professional or academic audiences can sit at 10th to 12th without losing readers.
Tools I’ve personally used at some point to help with all this include Hemingway, Grammarly, Yoast, and Readable.com. Readable is maybe my favorite since it gives you scores for all the formulas I’ve talked about here.
From there, just practice rewriting text if your readability score is too high. This article, for what it’s worth, sits at around a 7th- or 8th-grade level (if you remove the part with all the numbers where I’m talking about how the formulas work). That’s usually a good place to be.
Readability is accessibility, and practice makes perfect.
