Readability Score Calculator & Word Frequency Counter

This readability score calculator measures how easy your writing is to read, in bulk. Paste texts, give page URLs or pick a dataset and get one row per item with Flesch Reading Ease, Flesch-Kincaid grade, Gunning Fog, SMOG, Coleman-Liau and ARI, a consensus grade level, a plain-language reading level, word count, sentence count, syllable count, reading time and speaking time, plus a word frequency counter (top words with counts and percentages), lexical diversity and your longest sentences. It runs offline inside the Actor: no API key and no external service.

Pay only for items that got statistics. Empty, failed and skipped items are free.

What does this readability analyzer give you

• Seven readability formulas. Flesch Reading Ease (0-100+, higher is easier), Flesch-Kincaid grade, Gunning Fog, SMOG, Coleman-Liau, Automated Readability Index and an optional approximate Dale-Chall. A consensusGrade averages the grade-level formulas so you get one number to act on, and readingLevel gives a label such as "Standard (8th-9th grade)".
• Word counter and text statistics. wordCount, sentenceCount, syllableCount, charCount, charCountNoSpaces, paragraphCount, average words per sentence, average syllables per word, complex-word share.
• Reading and speaking time. Minutes to read silently (default 238 words per minute) and to read aloud (default 150 wpm). Both speeds are inputs.
• Word frequency counter. The top N words per text with count and percent, with an English stopword filter you can switch off and your own extra stopwords.
• Lexical diversity. uniqueWords, type-token ratio and the length-robust Guiraud index.
• Longest sentences. The longest sentences with their word counts, so you know what to split.
• Three input modes. One text per line, page URLs (main text extracted with trafilatura), or a field of another Actor's dataset.

How to calculate the readability score of a text

1. Paste your texts in Texts, one per line (the prefill has two examples, one plain and one dense).
2. Run the Actor.
3. Open the dataset overview table or download JSON, CSV or Excel. fleschReadingEase and consensusGrade are the columns to start with.

How to check the readability of a web page

Put the URLs in Page URLs. The article text is extracted with trafilatura and scored. The page title is returned. Public pages only, no login and no CAPTCHA solving. Pages that need JavaScript to show their text return empty with an explanation.

How to count word frequency in a dataset

Pick the Dataset and set Dataset field (default text, dotted paths and list positions work, for example review.text). Every string in that field becomes a row with its own top words. To get one frequency list for a whole corpus, join your texts into one entry first.

How to read the scores

Flesch Reading Ease	Reading level
90-100	Very easy (5th grade)
80-90	Easy (6th grade)
70-80	Fairly easy (7th grade)
60-70	Standard (8th-9th grade)
50-60	Fairly difficult (10th-12th grade)
30-50	Difficult (college)
10-30	Very difficult (college graduate)
below 10	Extremely difficult (professional)

Most web content works best at Flesch 60-70 or a consensus grade of about 8.

Input example

{
  "texts": [
    "Readability is how easy a text is to read. Short sentences and plain words make a page easy to follow.",
    "Notwithstanding the aforementioned considerations, organizations should systematically evaluate their communications."
  ],
  "urls": ["https://en.wikipedia.org/wiki/Readability"],
  "topWords": 20,
  "removeStopwords": true,
  "longestSentences": 3,
  "includeDaleChall": false,
  "maxItems": 1000
}

Output example

{
  "index": 1,
  "source": "text",
  "textPreview": "Readability is how easy a text is to read. Short sentences and plain words make a page easy to follow.",
  "status": "analyzed",
  "wordCount": 20,
  "sentenceCount": 2,
  "syllableCount": 29,
  "charCount": 102,
  "readingTimeMinutes": 0.08,
  "speakingTimeMinutes": 0.13,
  "avgWordsPerSentence": 10.0,
  "fleschReadingEase": 74.02,
  "fleschKincaidGrade": 5.4,
  "gunningFog": 8.0,
  "smog": 8.8,
  "colemanLiau": 5.1,
  "automatedReadabilityIndex": 2.6,
  "consensusGrade": 5.3,
  "readingLevel": "Fairly easy (7th grade)",
  "englishLikely": true,
  "uniqueWords": 16,
  "lexicalDiversity": 0.8,
  "topWords": [{"word": "easy", "count": 2, "percent": 16.67}, {"word": "readability", "count": 1, "percent": 8.33}],
  "longestSentences": [{"words": 11, "text": "Short sentences and plain words make a page easy to follow."}],
  "note": "short text (under 30 words): readability scores are unstable"
}

Use cases

• SEO and content teams: score every blog post or landing page and flag the ones above grade 10.
• Plain-language and accessibility audits: government, health, legal and finance copy.
• Education: check that reading material matches a grade level.
• Dataset enrichment: add readability and word counts to scraped reviews, articles or comments.
• Writers and editors: find the longest sentences and the most overused words.

How much does it cost

Pay per event: text-analyzed = $0.0008 per text, page or dataset item that received statistics ($0.80 per 1,000). Empty, failed and skipped items are not charged, and maxItems caps your spend. There is no start fee. The Actor is light (512 MB) and analyses thousands of texts per minute; page URLs add download time.

Honest limits

• Readability formulas are English formulas. Syllables are estimated with a rule-based heuristic (no dictionary), usually within a few percent, so scores can differ by a point or two from other tools. Numbers count as one syllable. Proper nouns are not counted as complex words in Gunning Fog and SMOG.
• Other languages: word count, sentence count, character counts, reading time and word frequencies work in any language with spaces between words (and per character for Chinese and Japanese). For Latin-script text that does not look English, the grades are still calculated but englishLikely is false and a note says to treat them as unreliable. For non-Latin scripts the grades are null. The built-in stopword list is English only; add your own with Extra stopwords.
• Short texts (under 30 words) give unstable scores and carry a note. SMOG is designed for 30+ sentences and joins the consensus grade from 10 sentences. Lists and headings without punctuation are treated as separate sentences only when separated by a blank line.
• Dale-Chall (lite) uses a rare-word proxy instead of the official 3,000-word list and is off by default.
• Page URLs: JavaScript-only pages, login walls and PDFs are not supported; pages are fetched with normal requests and no bot-protection evasion.
• Texts longer than Max characters analysed (default 200,000) are cut, and a note is added.

FAQ

What is a good Flesch Reading Ease score? 60-70 is plain English that most adults follow easily. Higher is easier. Technical, legal and academic writing often scores 20-50.

Why do I get different scores than another tool? Tools differ in how they split sentences and count syllables. The formulas here are the standard published ones; the inputs are estimated by documented rules. Compare scores from the same tool over time rather than across tools.

Does it work for languages other than English? Counts, reading time and word frequency do. The grade formulas were built for English, so for other languages they are flagged and should not be relied on.

Can I get word frequency for a whole corpus? Each row has its own top words. Join your texts into a single entry to get a combined list.

Are numbers counted as words? Yes, in wordCount and the readability formulas (one syllable each). They are left out of the word frequency list.

What does it do when something fails? One bad URL or text never stops the run. The row gets a status of failed, empty or skipped with an error, is not charged, and the run's SUMMARY record lists the totals.

Related tools

• Keyword Extractor: Extract Keywords & Keyphrases from Text for the key phrases of your texts.
• Language Detector: Detect Text Language in Bulk (API) to find out which language a text is in before scoring it.