HomeToolsUX writingReadability checker
Readability checker with the Flesch score
Paste your text — the checker computes Flesch Reading Ease and the Flesch–Kincaid grade level, estimates reading and speaking time, and highlights the longest sentences and words. Your text never leaves the browser.
—
Highlighted text
| # | Words | Sentence start |
|---|
Everything is calculated in your browser — nothing you enter is sent anywhere.
How to use it
- Paste your textAn article, newsletter, product description or interface copy. You need about 100 words for a stable result; on a couple of sentences the score jumps around.
- Read the scoreThe scale runs from 0 to 100: higher is easier. Next to it are the grade level, reading time and the main counts.
- Find the heavy partsThe longest sentences are highlighted most strongly, and words of three or more syllables are underlined. You can change the long-sentence threshold.
- Simplify and recheckSplit long sentences and swap long words for short ones — the score updates instantly as you edit.
The Flesch Reading Ease formula
The Reading Ease score was proposed by Rudolf Flesch in 1948. The idea is simple: text is harder when its sentences are longer and its words have more syllables.
The second formula is the Flesch–Kincaid Grade Level, developed in 1975 for the US Navy: it converts difficulty into a US school grade.
Worked example
Take three sentences: “Onboarding helps a new user quickly understand why they need the product. A good tooltip appears at the right moment and talks about one action. If people do not find value in the first minutes, they may close the tab and never come back.” That is 44 words, 3 sentences and 60 syllables.
A score of 77 is “fairly easy”, about sixth grade. English syllables cannot be counted exactly by rule, so the checker uses a common heuristic: it counts vowel groups and adjusts for a silent final “e” and “-ed”/“-es” endings. It gets most words right and is off by a syllable on some (“created”, “rhythm”), which shifts the score by a point or two on a typical text.
How to read the result, and where the score falls short
The checker uses Flesch’s original interpretation scale:
| Score | Rating | What it resembles |
|---|---|---|
| 90–100 | Very easy | Short sentences of simple words |
| 80–90 | Easy | Conversational text, simple instructions |
| 70–80 | Fairly easy | Good interface copy and newsletters |
| 60–70 | Plain English | Mainstream journalism, clear articles |
| 50–60 | Fairly difficult | Business correspondence, analysis |
| 30–50 | Difficult | Popular science and professional texts |
| 0–30 | Very difficult | Academic, legal and technical documents |
The score is useful for comparing versions of the same text and for finding heavy passages. But it is a crude model with known limitations:
- It does not see meaning. Short sentences of unfamiliar jargon score high, and meaningless strings of short words score perfectly.
- It ignores word familiarity. “Information” has four syllables but is familiar; “churn” has one but not everyone knows it.
- It is unstable on short texts. In a two- or three-sentence passage one long sentence moves the score by dozens of points. For interface copy the tooltip copy checker is a better fit.
- It depends on sentence splitting. Lists, headings and abbreviations with periods (“e.g.”, “i.e.”) confuse the count. The checker recognizes common abbreviations and decimals, and treats each line as a separate sentence.
- Grades are approximate. The formulas were calibrated on instructional material; for fiction or specialist documentation the grade mapping is rough.
Reading time: where 228 words per minute comes from
Reading time assumes an average speed of 228 words per minute. That figure comes from the multilingual IReST study (International Reading Speed Texts), in which adult native speakers of 17 languages read standardized texts silently; 228 wpm is the English average.
Speed varies many times over between people and texts: familiar topics read faster, instructions full of numbers and terms slower, and an interface screen is scanned rather than read. Speaking time is a rough estimate for a calm presentation pace.
For interface copy, time matters especially: people read a tooltip or tour step in passing, and anything that does not fit in a few seconds gets skipped.
How to make text easier to read
- Split long sentences. One idea per sentence. A sentence of more than 20 words can almost always be cut in two without losing meaning.
- Prefer verbs to nouns. “Make a selection” → “select”; “perform an update” → “update”. The text gets shorter and livelier.
- Cut filler phrases. “In order to”, “with regard to”, “at this point in time” add length and nothing else.
- Pick the shorter synonym. “Utilize” → “use”, “approximately” → “about”, “commence” → “start”.
- Use digits for numbers. “3 steps” scans faster than “three steps”.
- Recheck after editing. The score should rise, but the goal is clear text, not a record score.
If you write onboarding copy, start with the user onboarding checklist and SaaS onboarding examples, which cover how to word tooltips and steps.
Sources
- Flesch R. A New Readability Yardstick. Journal of Applied Psychology, 32(3), 1948 — the Reading Ease score.
- Kincaid J. P., Fishburne R. P., Rogers R. L., Chissom B. S. Derivation of New Readability Formulas (Automated Readability Index, Fog Count and Flesch Reading Ease Formula) for Navy Enlisted Personnel. Research Branch Report 8-75, 1975.
- Trauzettel-Klosinski S., Dietz K., IReST Study Group. Standardized Assessment of Reading Performance: The New International Reading Speed Texts IReST. Investigative Ophthalmology & Visual Science, 53(9), 2012 — reading speed in 17 languages.
- Plain Language Action and Information Network. Federal Plain Language Guidelines — plainlanguage.gov.
FAQ
What is the Flesch Reading Ease score?
A formula that estimates text difficulty from average sentence length in words and average word length in syllables. The result is a number from roughly 0 to 100: the higher, the easier. It was proposed by Rudolf Flesch in 1948.
What is a good readability score?
For mass-audience text — articles, newsletters, product pages — aim for 60 or higher. Interface copy and instructions should be easier still. Academic and legal texts rarely score above 30–40, and that is fine for their audience.
How do I calculate reading time?
Divide the word count by reading speed and multiply by 60 to get seconds. The checker uses 228 words per minute, the average adult silent reading speed for English in the IReST study.
Why is my score above 100 or below 0?
The formula is not bounded by the scale. Very short sentences of one-syllable words score above 100, and long sentences of long words go negative. On the scale the value is pinned to the edge.
How are syllables counted?
With a heuristic: vowel groups (including y) are counted, and a silent final “e” and some “-ed” and “-es” endings are subtracted. It is right for most words and off by one on some, which barely moves the score on a text of 100+ words.
Is my text sent to a server?
No. The analysis runs as JavaScript in your browser; the text is not transmitted or stored. Only the chosen long-sentence threshold is saved in the page URL.
How is Flesch–Kincaid different from Flesch Reading Ease?
It uses the same two measures with different coefficients and expresses difficulty as a US school grade: 5 is fifth grade, 9 is ninth grade. Values above 12 correspond to college-level text.
More in UX writing
- Tooltip copy checker
Is this tooltip short, clear and actionable enough?
- Tooltip & tour reading time
Do users have enough time to read a tooltip or tour step before it moves on?
- Email subject line preview
Will the subject line and preheader fit in the inbox on mobile and desktop?
- SERP snippet preview
How will the title and description look in Google, and will they be cut off?