Flesch Reading Ease Score: What the Number Means

The Reading Time & Readability tool reports an estimated reading time and a Flesch Reading Ease score for any text you paste in. The reading time is easy to understand. The score looks arbitrary until you know what it is measuring, and it becomes much more useful once you also know what it ignores.

What the Flesch Reading Ease score is

Rudolph Flesch published the formula in 1948 as a way to estimate how difficult a passage is to read, using two things that can be counted mechanically:

• Average sentence length, measured in words per sentence.
• Average word length, measured in syllables per word.

The formula is:

206.835 − 1.015 × (words ÷ sentences) − 84.6 × (syllables ÷ words)

Longer sentences subtract more. Longer words subtract much more, since the syllable term carries a bigger weight. The scale typically runs from 0 to 100, with higher meaning easier. It is not capped, so very short words in very short sentences can score above 100, while dense legal prose can dip below zero.

What the score ranges mean

The commonly cited bands are:

• 90–100: very easy; roughly fifth-grade level.
• 80–90: easy.
• 70–80: fairly easy.
• 60–70: plain English, understood by most adults and teens.
• 50–60: fairly difficult.
• 30–50: difficult; roughly college-level reading.
• 0–30: very difficult; specialist, academic or legal writing.

Most web content aimed at a general audience sits comfortably between 60 and 70. Technical documentation for engineers can reasonably sit lower. A children's story should sit much higher. There is no universally correct number, only a number that matches your readers.

How to check your text

1. Open Reading Time & Readability.
2. Paste a paragraph, a page or a full draft into the text box.
3. Read the three results: estimated reading time, Flesch Reading Ease score and the descriptive level.
4. Edit the draft: split your longest sentences and swap long words for shorter ones where the meaning survives.
5. Paste the new version and compare. The change in score is more informative than the absolute value.

Reading time is the word count divided by 200 words per minute, so a 1,000-word article shows five minutes. Real speeds vary with the reader and the material, so use it as a ballpark, for example to add a "5 min read" label. If you need exact counts of words, sentences and characters first, the Word Counter gives them.

Benefits of checking readability

• A fast, objective second opinion. Authors are poor judges of their own prose. A number flags paragraphs that deserve another look.
• A way to compare drafts. Did the rewrite make the text denser or lighter? The score answers that in a second.
• Consistency across a team. Pick a target range for help articles and keep every writer within it.
• Accessibility. Plain language helps readers with dyslexia, with limited attention, or who are reading in a second language.

What the formula deliberately ignores

Flesch only counts sentence length and syllables. It has no concept of vocabulary difficulty, jargon, ambiguity, structure or whether the ideas themselves are complex. "Notwithstanding the aforementioned" scores worse than "Use the second form", but a short sentence like "The party of the first part shall indemnify the party of the second part" is a mouthful of legal meaning that a short-word score will understate.

The syllable count itself is an approximation. The tool counts vowel groups and adjusts for a silent final "e", which is accurate on most English words but wrong on some: it will miscount words like "queue", "area" or "business" occasionally. Treat the score as accurate to within a few points, not exact.

Where typical readability tools fall short

Many readability checkers live inside paid writing platforms, require an account, or send your draft to a server for analysis, which is awkward when the text is an unpublished article, a client document or an internal policy. Some reduce everything to a single grade with no explanation of how it was reached. The MergeDoc tool runs in your browser: nothing is uploaded, there is no sign-up, and the formula is the standard one shown above. Its limits are honest ones. It is English-only, it uses a heuristic syllable counter, and it reports only the Flesch score and reading time rather than a full suite of metrics.

Everyday uses

• Blog posts and web copy. Check that a landing page reads at a plain-English level before publishing.
• Emails to customers. Run a long announcement through it and break up whatever scores lowest.
• Student essays and reports. See whether the draft is wordier than the assignment audience needs.
• Policies and terms. Find the clauses that are hardest going and rewrite them first.
• Spotting repetitive vocabulary. Pair it with the Word Frequency Analyzer to catch overused words.

Tips and mistakes to avoid

Do not chase a number. Chopping every sentence in half to inflate the score produces staccato writing that is tiring to read. Vary sentence length on purpose: a long sentence that builds an idea, followed by a short one that lands it, reads better than a uniform rhythm.

Score in chunks. A single score for a 5,000-word document hides its worst section. Paste in a section at a time to find the hard parts.

Remove headings, code samples and lists before measuring if they are not real prose. Fragments without full stops can be counted as one enormous sentence and drag the score down.

Know your audience. A 45 is fine for a specialist journal and a problem for a how-to guide for beginners.

Finally, read the text aloud. If you stumble, the formula may not have noticed, but your readers will.

A quick worked example

Take the sentence: "The cat sat on the mat." That is 6 words, 1 sentence and 6 syllables, so words per sentence is 6 and syllables per word is 1. The score comes out at 206.835 − 6.09 − 84.6 = about 116. That is above the usual 0 to 100 range, which shows why the scale is a guide and not a hard scale.

Now try: "Notwithstanding the aforementioned provisions, the undersigned acknowledges responsibility." That is 8 words, 1 sentence and about 27 syllables. Words per sentence is 8 and syllables per word is about 3.4, so the score is 206.835 − 8.12 − 285.5 = about −87. Same grammar, completely different feel, and the score captures the difference.

How to raise a low score without ruining the writing

• Split sentences at conjunctions such as "and", "but" and "which" when each half can stand alone.
• Replace long Latinate words with short ones when the meaning is unchanged: "use" for "utilize", "help" for "facilitate", "start" for "commence".
• Cut filler phrases like "in order to" and "it is important to note that".
• Keep the hard words you genuinely need. A technical term your audience knows is not a defect.

Re-run the tool after each pass and stop when the writing reads naturally, not when the number peaks.

Finally, remember that readability is relative to the reader. The same score can be perfect for a technical audience that knows the vocabulary and poor for newcomers who do not. If you can, test a draft on one real person from your target audience and ask where they slowed down. Their answer beats any formula, and the score then helps you check that the fix worked.

Related reading

If you write about tools and technical topics, Minified JSON Debugging is a good example of a how-to written for a general technical audience. For another kind of timed work, see The Pomodoro Technique: Why 25 Minutes Works. And if your writing includes numbers worth checking, Color Contrast and WCAG covers the other half of readability: whether the text can actually be seen. The text tools category has more.