Readability scores get quoted as if they were a ranking factor. They are not. They are a proxy for something that does matter - whether people finish reading - and understanding the formula tells you exactly which edit fixes a bad score.
The formula
Flesch Reading Ease, 1948, still the most used:
206.835 − 1.015 × (words / sentences) − 84.6 × (syllables / words)
Two inputs: average sentence length, and average syllables per word. Higher output is easier.
The coefficients matter. Look at the practical weight of each term: reducing average sentence length from 25 to 15 words adds about 10 points. Reducing average syllables per word by a tenth adds about 8 - and cutting a tenth of a syllable off your average vocabulary is a much harder edit than splitting sentences.
So the fastest way to improve readability is to split long sentences. Not to swap words for shorter synonyms, which usually makes technical writing worse.
Free toolReadability CheckerScore your writing on Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG and Coleman-Liau at once, and see which one your text fails.What the bands mean
| Score | Level | Typical of |
|---|---|---|
| 90–100 | 5th grade | Children's books |
| 80–89 | 6th grade | Consumer marketing |
| 70–79 | 7th grade | Popular fiction |
| 60–69 | 8th–9th grade | Plain English, most journalism |
| 50–59 | 10th–12th grade | Trade press |
| 30–49 | University | Technical documentation |
| 0–29 | Postgraduate | Academic papers |
Aim for 60–70 for general audiences. Technical content for developers naturally sits at 30–50 and that is appropriate - forcing developer documentation to grade 8 makes it longer and vaguer, not clearer.
Does it affect rankings?
Not directly. Google has never confirmed a readability signal and there is no evidence of one in the ranking systems.
It affects rankings indirectly through behaviour. A page people bounce off in ten seconds does not accumulate the engagement signals that do matter. If your content is genuinely hard to read, fewer people finish it, fewer link to it, and fewer come back.
The honest summary: readability is a writing-quality diagnostic, not an SEO lever. Treat a bad score as a symptom to investigate, not a number to optimise.
What the score misses
It counts syllables and sentence lengths. It has no idea whether the writing is any good.
- Jargon is invisible to it. "We leverage synergistic paradigms" scores fine.
- Structure is invisible. A wall of text and a well-broken-up page with the same sentences score identically, but only one is actually readable.
- Order is invisible. Burying the answer in paragraph six costs you readers and costs you nothing on the score.
- Short can be worse. Chopping every sentence to eight words scores beautifully and reads like a ransom note.
The things that most improve real readability - putting the answer first, using descriptive subheadings, replacing a paragraph with a table - do not move the score at all.
The edits that work, in order
- Split any sentence over 25 words. Usually at a conjunction. Biggest single win.
- One idea per paragraph. Three or four sentences maximum.
- Answer first. Lead the section with the conclusion, then explain.
- Use a table when you are comparing things. Prose comparisons are hard to read at any score.
- Cut hedges. "It is important to note that" and "in order to" add syllables and remove nothing.
- Only then look at vocabulary - and keep every technical term your reader needs.
Run the score once before publishing, as a smoke test. If it comes back under 30 for non-academic content, the problem is almost certainly sentence length, and splitting a dozen sentences will fix it.
Launch it where the numbers are checked
RankCert ranks products on domain control we verify ourselves. Listing is free and the link stays dofollow whether or not you display the badge.