To check readability score in Grammarly, copy the English passage into the Readability Score Checker and select Analyze to calculate six published readability measures at once. This method evaluates the copied text; it does not claim to reproduce a score or menu inside Grammarly. Use it to inspect both the result and the counts behind it for a draft, email, article, or other English passage. The checker presents Flesch Reading Ease, Flesch-Kincaid Grade Level, Gunning Fog, SMOG, Coleman-Liau, and the Automated Readability Index together. It also reports words, sentences, syllables, letters, characters, three-syllable complex words, and polysyllables. Start with the Flesch Reading Ease band, compare the five U.S. grade-level outputs, and inspect the raw counts before revising. The formulas are expected to disagree because they weigh different signals. That disagreement is useful when checked against the underlying counts instead of treating one number as a verdict. For unpublished or sensitive drafts, the analysis runs in your browser, so the text is not uploaded, stored, or attached to an account. Readability formulas remain directional: they can show that an edit made a paragraph denser, but cannot guarantee that a particular reader will understand it.
How to check readability score in Grammarly drafts
- Copy and paste the target text. In Grammarly, select the English paragraph, email, or article section you want to assess, copy it, and paste it into the checker’s text box.
- Calculate all six measures. Select Analyze. The tool evaluates the pasted passage with Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG, Coleman-Liau, and ARI at the same time.
- Read the output in order. Begin with the Flesch Reading Ease band, compare the five grade-level scores, and then check the raw words, sentences, syllables, letters, characters, complex words, and polysyllables underneath.
For a before-and-after review, analyze the original passage, make one focused revision, and paste the revised text into the box again. Keep the passage boundaries and content type consistent between runs so that changes in the counts reflect your edits rather than a completely different sample.
Read the six outputs on the correct scales
The six results are related, but they are not interchangeable. An ease score and a school-grade score answer different questions.
| Measure | Output scale | Main counting signal | Useful role |
|---|---|---|---|
| Flesch Reading Ease | 0-to-100-style ease score; higher is easier, while values may exceed 100 or fall below 0 | Words, sentences, and syllables | Start with its plain-language interpretation band |
| Flesch-Kincaid Grade Level | U.S. school grade | Words, sentences, and syllables | Estimate grade-level demand while checking syllable sensitivity |
| Gunning Fog | U.S. school grade | Long words relative to sentences | Highlight prose weighted toward difficult words |
| SMOG | U.S. school grade | Polysyllables and sentence count | Cross-check syllable-heavy prose, with special caution on short samples |
| Coleman-Liau | U.S. school grade | Letters or characters relative to words and sentences | Provide a character-based cross-check |
| Automated Readability Index | U.S. school grade | Characters relative to words and sentences | Compare with Coleman-Liau when syllables may be estimated incorrectly |
Flesch Reading Ease follows a 0-to-100-style scale only in appearance. Its score is genuinely unbounded, so the checker uses open-ended interpretation thresholds instead of treating 0 and 100 as hard limits. Higher values indicate easier prose. The band labels follow the convention in the MSKTC plain-language document on Flesch Reading Ease bands, not wording attributed to Flesch himself.
The other five formulas return U.S. school-grade levels, so they should be read on a grade scale rather than compared directly with the Reading Ease number. A higher grade represents a greater level of demand. Coleman-Liau and ARI can also return a negative grade for a very short, very simple sentence. That is the published arithmetic, not necessarily a processing error.
Why the six scores naturally disagree
Disagreement is not automatically a sign that an analysis failed. Flesch-Kincaid and SMOG depend strongly on syllables, while Gunning Fog gives particular weight to long words per sentence. Coleman-Liau and ARI use characters rather than syllables. Those character-based results provide a useful cross-check when the estimated syllable count looks questionable. Reading all six together is therefore more informative than declaring one formula the universal winner.
The implementation checks its constants against primary or authoritative material rather than copying popular blog tables. The 1975 Navy technical report that derived the Flesch-Kincaid and ARI formulas supplies those coefficients; McLaughlin’s original 1969 paper supplies the SMOG constants; and the MSKTC document supports the Flesch coefficient and interpretation bands. The Flesch coefficient is 84.6, not the 83.6 that an OCR scan appears to show. Coleman-Liau uses 0.0588, not the 0.0855 found in at least one circulating journal table. When published sources conflict, the checker resolves the value with independent evidence rather than counting search results.
The displayed counts provide the audit trail. Readability scores are functions of inputs such as words, sentences, syllables, and characters. When two tools produce different results for identical text, they often disagree about how those inputs were counted rather than about the underlying formula.
Use the raw counts as an audit trail
Pay attention to the numbers beneath the scores:
- Words and sentences show the overall sample size and the sentence-length inputs used by the formulas.
- Syllables affect the Flesch, Flesch-Kincaid, and SMOG calculations.
- Letters and characters provide the inputs used by Coleman-Liau and ARI.
- Three-syllable complex words and polysyllables expose the long-word signals used by the relevant formulas instead of hiding them inside a total.
The syllable count is a deterministic estimate because English has no authoritative algorithm for deriving syllables from spelling. The checker splits hyphenated words, handles consonant-plus-le endings, decides whether a final “es” or “ed” adds a syllable, drops silent final “e,” and counts the remaining vowel groups. Every word has a floor of one syllable. These rules are tested against a list of hand-checked words and work well on ordinary prose, but they can undercount some vowel-adjacent words.
Sentence detection is also heuristic. Decimals, initials, and a short abbreviation list are shielded so that a price, initial, or title does not appear to end a sentence early. If a grade-level result surprises you, inspect the sentence and syllable totals before rewriting the passage. The counts may explain the formula more clearly than the final number does.
Recognize the results that need extra caution
Short text and the SMOG caveat
McLaughlin defined SMOG for a sample of exactly thirty sentences. Below that length, the checker still displays the score but marks it as an extrapolation from a smaller sample. Because the formula includes a square root, its scaling can also amplify counting noise. Treat a SMOG result from a brief paragraph as an estimate rather than a stable measurement.
Lists, code, tables, and numbers
Grade-level formulas were developed for running prose. Results for lists, code, tables, or text dominated by numbers are indicative only because those structures do not represent ordinary sentence and word patterns. Use their numerical features for editing, but do not treat the resulting grade as a reliable audience rating.
A score is a direction, not a verdict
Readability measures identify textual features; they do not measure every aspect of comprehension or predict a specific reader’s reaction. A score can reveal that a paragraph became denser after an edit, but you still need to consider context, organization, terminology, and the intended audience.
Turn the report into a focused revision
- Start with representative prose. Analyze a section that reflects the running text you want to simplify rather than a page of mixed headings, lists, and tables.
- Read the ease band first. Use Flesch Reading Ease for a quick orientation, not as the sole measurement.
- Compare the grade results. Look for agreement and disagreement among the five grade-level formulas instead of averaging them into a new, unsupported score.
- Trace unusual results to the counts. Check whether long sentences, multisyllabic wording, or character-heavy terms are driving the difference.
- Revise one factor and analyze again. Compare similar passage lengths, then read the revision manually for clarity.
The practical answer to how to check readability score in Grammarly is therefore straightforward: copy the relevant English text, paste it into the checker, select Analyze, and read the band, grade levels, and counts together. This provides a transparent reading-level check for a Grammarly draft without presenting an independent formula as a native Grammarly metric or as a guarantee of comprehension.