Text Readability Calculator

Introduction to English text readability scoring

This text readability calculator estimates how demanding an English passage may feel by counting measurable features: words, sentences, syllables, letters, characters and longer words. It reports six established indices rather than claiming that one score can describe every reader, purpose or subject. Use it when revising public information, classroom material, instructions, policy drafts, reports, emails or web copy. The most useful question is often whether a revision is clearer than the earlier draft and whether its level suits the people who need to act on it.

Readability is not the same as comprehension. These equations cannot know whether an explanation is accurate, logically organised, culturally familiar, visually well designed or supported by a useful heading, example or diagram. They can nevertheless reveal patterns worth reviewing. A long sentence may hide several decisions in one line. Dense vocabulary may be essential in specialist writing, but it can need a definition when the audience is new to the topic. The calculator therefore displays the source counts and the sentences with the highest estimated Flesch-Kincaid levels.

How to use this readability analyser on a draft

Paste a representative passage into the field below, choose the reading grade that best reflects the intended audience, and select Analyse readability. A sample of 100 words or more is a practical minimum because a very short passage can change sharply when one sentence is added or removed. Several hundred words drawn from the actual page, letter or instruction set are better. Keep headings, bullets and punctuation in the form readers will encounter. Add terminal punctuation to list items when you want them counted as separate sentences.

The result begins with Flesch Reading Ease and a consensus grade. The consensus is the median of five grade-level measures, so one unusually high or low formula has less influence than it would in a simple average. Compare that median with the chosen target, then inspect the individual indices and the counts. Re-run the calculation after a meaningful edit rather than changing words solely to chase a number. Copy summary saves a concise result to the clipboard, while Download CSV creates a small record of the counts and scores for a revision log.

Readability formulas used by this calculator

Let be the number of words, the number of sentences and the number of syllables. Average sentence length and average syllables per word are the central ratios in the syllable-based formulas:

Formula: ASL = W / S, ASW = Y / W

Flesch Reading Ease formula

Flesch Reading Ease is a roughly 0-to-100 scale on which higher values indicate easier prose. It is particularly useful for comparing versions of similar public-facing text. Scores can fall below zero for unusually long and syllable-heavy passages, so it should be read as a comparative scale rather than a percentage of readers who will understand the document.

Formula: RE = 206.835 - 1.015 ⁢ W / S - 84.6 ⁢ Y / W

Flesch-Kincaid Grade Level formula

Flesch-Kincaid uses the same two ratios but expresses the output as an approximate United States school grade. Its direction is intuitive: larger grade values mean longer sentences, more syllables per word, or both. It is a useful editing signal, not a test of intelligence, eligibility or a reader’s personal capability.

Formula: FKGL = 0.39 ⁢ W / S + 11.8 ⁢ Y / W - 15.59

Gunning Fog Index formula

Gunning Fog combines sentence length with the percentage of complex words. Here is the complex-word count. The implementation excludes likely proper nouns and hyphenated compounds from this count and attempts to avoid treating ordinary inflectional endings as added complexity. That convention makes Fog helpful for spotting dense prose, while reminding writers that a technical term may still be necessary.

Formula: Fog = 0.4 ⁢ W / S + 100 ⁢ C / W

SMOG grade formula

SMOG uses polysyllables, including repetitions and proper nouns, and scales the count to a 30-sentence basis. Its output should be read carefully on a short sample because that scaling becomes an extrapolation. With as the polysyllable count, a word contributes when its estimated syllable total is at least three:

Formula: P = count(syllables ≥ 3)

Formula: SMOG = 1.0430 ⁢ sqrt(P ⁢ 30 / S) + 3.1291

Coleman-Liau Index formula

Coleman-Liau does not require syllables. It uses , letters per 100 words, and , sentences per 100 words. This can make it differ from syllable-based formulas when words are short in letters but have more syllables, or the reverse. Its two rate definitions are:

Formula: L = 100 ⁢ letters / W

Formula: S ′ = 100 ⁢ S / W

Formula: CLI = 0.0588 ⁢ L - 0.296 ⁢ S ′ - 15.8

Automated Readability Index formula

The Automated Readability Index, or ARI, also avoids syllable estimation. It uses for alphanumeric characters, together with words and sentences. Because it weights characters differently, ARI is useful as a second perspective rather than a replacement for the other formulas. Character-heavy terminology can increase ARI even where a spoken syllable count seems moderate.

Formula: ARI = 4.71 ⁢ Ch / W + 0.5 ⁢ W / S - 21.43

How the readability calculator counts text

Words are runs of letters, optionally joined by an apostrophe, typographic apostrophe or hyphen; numeric strings such as 3.14 are also retained. Sentence boundaries are detected at full stops, question marks and exclamation marks when the following character permits a boundary. The counter protects common abbreviations, initials and decimal points where possible. If the text has words but no ending punctuation, the whole passage is treated as one sentence. The result explains why that treatment can overstate difficulty.

Syllables are estimated by a vowel-group heuristic with common exceptions. It accounts for silent final e, consonant-plus-le, many -ed and -es endings, initial y and selected irregular words. This creates consistent editing feedback, not a pronunciation dictionary. Names, loanwords, technical terms and ambiguous vowel sequences can still be miscounted. Treat a decimal result as a comparison tool, and check unfamiliar terms manually before making an important decision.

Counts also need context. A heading, web address, citation, product code or table may not behave like ordinary prose. A short sentence made mostly of names can look difficult even when its meaning is obvious to the intended reader. Conversely, a low score can hide an inaccurate instruction or a missing condition. The calculator is strongest on connected English prose with normal punctuation, where it can help a writer find sections that deserve a closer human review.

Worked example: a short clinic reminder

Consider this reminder: “The clinic will call you before your appointment. Bring a list of your medicines. If you cannot come, tell us at least one day early.” It has short sentences and familiar verbs, but words such as appointment and medicines add syllables. The calculator will likely show an easy Flesch score and a noticeable spread among the grade indices because the sample is only three sentences long. That spread is a prompt to collect a longer sample, not evidence that the reminder is unsuitable.

For illustration, if the passage has , and , then and . Substituting those ratios into Flesch Reading Ease produces:

Formula: RE = 206.835 - 1.015 ⁢(8.333) - 84.6 ⁢(1.320) = 86.7

The equivalent Flesch-Kincaid calculation is:

Formula: FKGL = 0.39 ⁢(8.333) + 11.8 ⁢(1.320) - 15.59 = 3.2

The practical edit is not necessarily to replace every longer word. “Appointment” is likely the clearest accurate word in a clinic reminder. Instead, preserve the necessary term, make the requested action direct, and give readers enough time information to respond. A longer patient handout could define an unavoidable medical term once, then use consistent language throughout. This supports understanding even if a formula changes only slightly.

Interpreting readability results for the intended audience

A Flesch Reading Ease result around 60 to 70 is commonly described as standard or fairly easy general prose; higher scores are easier on that scale. Grade values are not admissions requirements and are not universal measures of intelligence. A grade 6 public-health handout, grade 8 consumer page and grade 12 professional explanation can all be appropriate when their vocabulary, detail and audience differ. The best target is usually the lowest level that preserves the precision the reader needs to make a sound decision.

When the formulas cluster, the median is a useful quick summary. When they differ by several grades, the pattern itself is useful. Higher Fog and SMOG with lower ARI often points to short sentences containing polysyllabic terms. Higher ARI and Coleman-Liau with lower Fog can point to long words in letters or longer sentence structures. Before simplifying terminology, consider adding a plain-language definition, an example, a visual cue or a clearer sequence of actions. Those changes may improve real understanding even when a formula barely moves.

Use the sentence list as a starting point for an editorial conversation. Ask who performs the action, what they must do first, what condition changes the instruction, and what outcome follows. Split a genuinely overloaded sentence when it contains more than one action. Prefer specific everyday verbs where they remain accurate. Do not turn complete sentences into fragments just to lower a score, because abrupt fragments can reduce clarity, accessibility and trust.

Limitations of readability formulas and count estimates

These formulas were developed from particular English-language samples and reader tests, not every modern genre. They are less reliable for code, URLs, citations, tables, poetry, product names, mixed-language text and heavily formatted lists. They also cannot judge layout, prior knowledge, tone, factual accuracy, visual hierarchy or whether a reader can find the needed instruction quickly. A passage can be made numerically easier by chopping every sentence into fragments and replacing precise terms with vague ones; that is not automatically better writing.

Use the output as an editing signal. Read difficult sentences aloud, identify the action and actor, break up genuinely overloaded sentences, define unavoidable technical terms and test important information with real users where possible. For high-stakes health, legal, financial or safety material, readability formulas are only one quality check among expert review, accessibility review and user testing. Pasted text stays in this browser while the calculator runs, but a local calculation does not remove the need to protect confidential material according to your organisation’s policies.

Questions writers ask about readability scoring