Readability Checker

Which score should you actually use?

Every formula here estimates difficulty from sentence length plus a measure of word difficulty. Four of them count syllables; Coleman-Liau and the Automated Readability Index count letters instead, which is why they disagree with the others on words like through. They also differ in how they weight the two. That is worth knowing, because it tells you what they can and cannot see.

What "good" looks like

Most plain-English guidance in the UK and US public sector targets a reading age of around 13 to 14 — roughly grade 8 to 9, or a Flesch ease of 60 to 70. That is not because readers are unintelligent. It is because people skim, read on phones, and are usually tired or in a hurry. Popular novels sit around grade 7. Broadsheet journalism sits around grade 10 to 12.

If you write about tax, medicine or law, you will not reach grade 8 without lying. Aim to be as clear as the subject allows, then check that every long sentence and every piece of jargon is doing real work.

The five things highlighted above

Passive voice

"The application was rejected" hides who rejected it. Passive constructions are the single biggest reason official writing feels evasive. They are not always wrong — when the actor genuinely does not matter, or is unknown, passive is the honest choice — but each one deserves a second look.

Adverbs

Most -ly adverbs prop up a weak verb. "Walked quickly" is usually "hurried". Cutting them tightens prose without losing meaning, which is why the highlight exists — not because adverbs are forbidden.

Filler and hedging

"Very", "really", "basically", "actually", "in order to". Delete them and the sentence almost always survives intact and reads stronger. Hedging is different from precision: "roughly 40%" is precise about its imprecision; "quite a lot" is not.

Clichés

Fourteen phrases that were vivid once and are now filler: at the end of the day, touch base, circle back, low-hanging fruit, moving forward, paradigm shift. They cost you words without adding meaning, and in an essay they read as padding to a marker who has seen them a thousand times. Cutting one almost never loses anything — the sentence is usually clearer with the phrase simply deleted rather than replaced.

Long sentences

Anything over 25 words is flagged. Long sentences are not automatically bad, but a document made entirely of them is exhausting. The fix is usually one full stop in the middle, not a rewrite.

What a score cannot tell you

No formula understands meaning. A page of short, simple, completely wrong sentences scores beautifully. These numbers measure the mechanics of difficulty and nothing else — treat a low score as a prompt to reread, never as a verdict on whether the writing is any good.

They also assume English prose. Bullet lists, tables, code and headings distort every formula here, so score the running text rather than the whole page.

One more honest caveat: syllables are estimated from spelling, the way every automated readability tool does it — counting a real dictionary would mean shipping one. Words with adjacent vowels come out a syllable short, so idea, area and science count as two, two and one. That nudges Reading Ease slightly up and the grade levels slightly down on text with a lot of them. It is a small effect across a paragraph and a real one across a single sentence.