Extract the most common bigrams, trigrams, 4-grams, or 5-grams in any text. Filter by minimum frequency and drop pure-stopword phrases.
Built and maintained by Paul Clark, Redmoon Software
Use the n-gram analyzer to find common phrases in writing, transcripts, or scraped content. Especially useful for SEO topical analysis, content gap analysis, and discovering author voice patterns.
The n-gram analyzer is more general than Keyword Density (no SEO framing). It's the natural next step after Word Frequency if you suspect important phrases lie above the single-word level.
An n-gram is a contiguous sequence of n words from a text. Bigrams are 2-word sequences ("machine learning"); trigrams are 3-word ("artificial neural network").
The analyzer counts every n-gram, filters by minimum frequency, and optionally drops phrases composed entirely of stopwords.
Lowercase normalization treats "The Cat" and "the cat" as the same n-gram — usually what you want for analysis.
Keyword density adds percentages and over-optimization warnings for SEO. The n-gram analyzer is a general phrase-frequency tool without the SEO framing.
Without filtering, the top n-grams are always "of the", "in the", "to the" — uninformative. The filter drops phrases composed entirely of stopwords.
For finding repeated long phrases — quotes, slogans, boilerplate. Most useful n is 2 or 3.
Input
500-word product description.
Output
Top bigrams: "machine learning" (8), "neural network" (5), "training data" (4).
N-grams reveal the actual phrases that dominate a text, including phrases the writer didn't consciously target. Useful for SEO topic discovery and content audit.
N-gram analysis counts sequences of adjacent words rather than individual ones, which surfaces a category of repetition that single-word counts miss entirely. A writer who uses "it is important to note that" nine times has not repeated any single word enough to stand out, but the phrase is doing visible damage.
Bigrams tend to surface genuine terminology — the fixed two-word terms a document depends on. Trigrams and longer surface habits and templates. That difference is why it is worth running both rather than picking one.
A repeated phrase is not automatically a problem, and the tool cannot tell which is which. Technical documents *should* repeat their terms exactly: renaming a component halfway through for variety is far worse than using the same noun forty times. Consistency of terminology is a virtue that looks identical to repetition in a frequency count.
The judgement is whether the phrase is a name or a habit. Names should repeat. Constructions — transitional formulas, throat-clearing openers, stock evaluative phrases — should not, and those are what to scan the list for.
% content words vs. function words.
Text analysisVocabulary variety — type‑token ratio.
Text analysisTotal + per‑word syllable counts.
Text analysisHighlight passive sentences in your writing.