Readability Score and Keyword Density Checker

Get Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau and ARI scores, reading time, sentence statistics, top keywords and phrases and exact target keyword density for text you provide. Bulk, deterministic, no web calls, no AI.

What you get

One record per text with readability scores, a flesch_label, structure counts, reading_time_minutes, top_keywords and target_keywords with counts and density.

Use cases

• Readability score check for blog posts, product pages and documentation before publishing.
• Keyword density report for SEO writers: count your target keywords exactly.
• Content QA at scale: flag texts with long sentences or a high grade level.

Analyse text you provide and get, in one record per text: readability scores, reading time, sentence statistics, the most frequent words and phrases, and the exact count and density of your target keywords. No web calls, no AI model, deterministic. Callable by developers and AI agents through Apify's MCP server (https://mcp.apify.com).

• Readability (English): Flesch Reading Ease, Flesch-Kincaid grade, Gunning Fog, SMOG, Coleman-Liau, Automated Readability Index, complex-word share.
• Structure: characters, words, unique words, sentences, paragraphs, average and longest sentence, number of sentences over your word limit.
• Reading time at 238 words per minute (Brysbaert 2019, silent reading of non-fiction: https://doi.org/10.1016/j.jml.2019.104047).
• Keywords: top words and 2-3 word phrases with counts and density (occurrences / total words x 100), optional English stopword filtering, plus your own target keywords counted as whole words (case-insensitive, Unicode aware, so it works for any space-separated script).

Input

Field	Meaning
texts	Texts to analyse (up to 2,000, each up to 500,000 characters)
topKeywords, maxPhraseLength, removeStopwords	keyword list options
targetKeywords	words or phrases to count exactly
longSentenceWords	threshold for "long sentence" (default 25)
readability	turn the formulas off if you only need keywords

Example input and output

Input (bundled test-input.json): one 84-word text, topKeywords: 8, maxPhraseLength: 2, targetKeywords: ["actor", "AI agent"]. Real output (abridged):

{
  "words": 84, "sentences": 6, "paragraphs": 2, "avg_words_per_sentence": 14,
  "longest_sentence_words": 20, "reading_time_minutes": 0.35,
  "readability": { "applicable": true, "flesch_reading_ease": 60.69, "flesch_kincaid_grade": 8.27,
                   "gunning_fog": 9.89, "smog_reliable": false, "automated_readability_index": 7.44 },
  "flesch_label": "standard (8th-9th grade)",
  "top_keywords": [ { "phrase": "actor", "words": 1, "count": 3, "density_percent": 3.571 } ],
  "target_keywords": [ { "keyword": "actor", "count": 3, "density_percent": 3.571 },
                       { "keyword": "AI agent", "count": 1, "density_percent": 2.381 } ]
}

Limitations (read before relying on it)

• Readability formulas are for English and are skipped (applicable: false) when fewer than 80% of words are Latin-script. Keyword counts work in other languages; there is no stemming, so actor and actors count separately, and the stopword list is English only.
• Syllables are counted by a heuristic with a small exception list, not a pronunciation dictionary. It is right for most common English words but can be off by one on some, so scores can differ by a few points from other tools. The formulas are the published ones (Flesch 1948; Kincaid et al. 1975; Gunning 1952; McLaughlin 1969; Coleman and Liau 1975; Senter and Smith 1967).
• The formulas are designed for passages of about 100+ words; shorter texts get a note. SMOG needs 30+ sentences (smog_reliable).
• Sentence splitting is rule-based: it protects decimals, initials and common abbreviations (Dr., Mr., U.S., etc.) but a sentence ending in an abbreviation such as "etc." is merged with the next one.
• No keyword density is "right". The Actor reports numbers; it does not claim a target percentage improves search ranking.
• Reading time is an average-reader estimate, not a measurement.

Pricing (pay per result)

Pay-per-event with Apify's built-in dataset-item event: each text is one result. The live price is set in Apify Console and shown on the Store page (see Store). Planning figure in our notes: USD 2.00 per 1,000 texts. Competitor snapshot (search results, checked 22 Oct 2026): readability-only Actors list from USD 0.50 to USD 10 per 1,000, and one at USD 250 per 1,000 (see apify/RESEARCH-3.md).

FAQ

Does it fetch web pages? No. Pass the text itself.

Does it use AI? No.

Does it handle HTML? No; strip markup first.

Which readability score should I use? Flesch Reading Ease is the most widely quoted (higher is easier); the grade-level formulas (Flesch-Kincaid, Fog, ARI) map to school years.

Does it support Arabic? Keyword counts do; readability formulas are English only.