What is N-gram analysis?

An n-gram is a run of n words: a unigram is one word, a bigram two, a trigram three. N-gram analysis counts every run in your manuscript and lists the ones you use most. Familiar counts runs of one to six words, chapter by chapter, so you see how often a word or phrase appears and where in the book it does.

How can N-gram analysis improve my writing?

It shows you habits that are hard to see from inside the book: a word you reach for in every scene, a phrase that comes back every few chapters, filler words like “just” and “felt”. A high count is not always a problem, since some repetition is a motif. The report gives you the counts and where they fall; what to change is up to you.

What should I look for in my word frequency results?

Start with the Crutches tab and the three words flagged at the top of the report. Then look for descriptive words and verbs that sit well above your other words, and for longer phrases that repeat stock descriptions. The bar for each chapter shows whether a word is spread through the book or bunched in a few chapters, which usually points to the scene to revise.

How does character dialogue analysis work?

Character dialogue analysis is not available yet. The closest thing today is style analysis, which reports each chapter’s point of view and tense and lets you compare how much of each chapter is dialogue.

Can editorial analysis help with genre-specific writing?

Familiar does not score your book against a genre benchmark. It reports your own rates per 1,000 words and per chapter, which you can set against books you know in your genre. Style analysis also flags stock phrases that are overused in contemporary genre fiction, such as “a pang of” and “the ghost of a smile”.

How much does editorial analysis cost?

N-gram analysis costs 1 credit and style analysis 2 credits. Credits are $5 each. Running every editorial analysis together with Analyze Everything costs 12 credits instead of 15.

For authors

Editorial analysisFind the words your book leans on

Familiar counts every word and phrase in your manuscript, chapter by chapter, and points out the crutch words, repeated phrases and rhythm habits an editor would mark. You decide what to change.

Get started Explore all tools

N-gram analysis 1 credit ($5)
Style analysis 2 credits ($10)
Results in a few minutes

N-gram analysis · 1 credit

Most-used words

moths · 212

orchard · 168

grandmother · 97

ashgrave · 84

hives · 61

whispering · 47

Take a second look

felt

Used 298 times, 3.1 per 1,000 words, peaking at 22 in chapter 7.

Filter verb

Chapter 1 · The Inheritance

The moths came to the window the night my grandmother died, and every one of them was whispering. I counted forty before I stopped counting. They pressed against the glass in the shape of her name, and when I opened the latch, not one of them came in.

What the analysis counts

Two reports read the whole manuscript. N-gram analysis counts words and phrases; style analysis looks at how the prose reads. For scores on plot and character, or the shape of the story, see manuscript evaluation and story beats.

Repeated words and phrases

See where a word clusters, not only how often it appears.

Familiar lists your most-used words, two- and three-word phrases, and longer phrases up to six words. Each one comes with its count, its rate per 1,000 words, and a bar for every chapter, so a word that piles up in two scenes stands out from one spread evenly through the book.

The top 100 phrases at each length, from one word to six
Words like “the” and “of” left out of single words and short phrases
Longer phrases keep them, to catch repeated beats and stock descriptions

N-gram analysis · repeated across chapters

PhraseCountPer 1kChapters 1–24lord ashgrave580.6whispering470.5moths came120.1yellow wax90.1the night my grandmother died60.1pressed against the glass50.1

Longer phrases

the night my grandmother died5 words · 6 times · chapters 1, 3, 10, 20 and 24

Repeated on purpose? Then it is a refrain. If not, it is a beat the book keeps replaying.

Crutch words

The filler words other word counts leave out.

Most word counts skip common words, which is where crutches hide. Familiar counts them separately: intensifiers like “very” and “really”, hedges like “just” and “maybe”, filter verbs like “felt”, “seemed” and “noticed”, and dialogue tags such as “nodded” and “hissed”.

The three most frequent flagged at the top of the report
Each with its rate per 1,000 words and the chapter where it peaks

Crutches

feltFilter verb298 · 3.1 per 1k

justHedge211 · 2.2 per 1k

seemedFilter verb142 · 1.5 per 1k

noddedDialogue tag74 · 0.8 per 1k

veryIntensifier66 · 0.7 per 1k

reallyIntensifier58 · 0.6 per 1k

Chapter 7 · felt, 22 times

I felt the cold come off the glass, and I felt them watching me from the dark.

Style analysis: rhythm and voice

Hear the rhythm, and catch the slips.

Style analysis measures sentence length and variety, finds runs of sentences that are all about the same length, and lists the words that open sentence after sentence. It also works out each chapter’s point of view and tense and shows the individual sentences that slip out of them, each checked in context before it is reported.

Narration only, so a quick exchange of dialogue is not read as choppy prose
Point-of-view and tense slips, each with a suggested fix

Style analysis · 2 credits

Sentence lengths

1–56–1011–1516–2021–2526–3031–4041+

Most common opener: “I”, 14% of narration sentences

Point of view & tense

First personPast tense23 of 24 chapters agree

Slip · Chapter 9

Tense

When the light ~~goes~~ went out the moths go went with it, all at once.

Style analysis: phrasing and drift

Find the chapter that reads like a different book.

Familiar compares each chapter’s word habits with the rest of the book and says why one stands apart: longer sentences, more dialogue, less varied vocabulary, more -ly adverbs. It also counts stock phrases and clichés, and marks passages an editor might look at again, such as awkward phrasing, vague detail or purple prose.

Chapter drift for books with six or more chapters
Stock phrases counted chapter by chapter
Passage notes that quote the page and say why

How far each chapter sits from the rest

Reads differently

Chapter 17

Dialogue share46% · others 19%

Sentence length (words)9.8 · others 15.2

Stock phrases

a pang of7

the ghost of a smile3

How it works

Add your manuscript
Create a book in Familiar and upload the manuscript. Chapters are found for you.
Run the analysis
Start n-gram analysis for 1 credit, style analysis for 2 credits, or run every analysis at once for 20% less.
Read it by chapter
Results are ready in a few minutes, with every count tied back to the chapters it comes from.
Revise and re-run
Upload your revised draft and run the analysis again on the new version.

Editorial analysis pricing

Pay per analysis with credits. Every plan includes credits each month, and extra credits are $5 each.

N-gram analysis1 credit · $5
Style analysis2 credits · $10
Analyze Everything (all 8 analyses, 20% off)12 credits · $60

Get started

Analyze Everything also includes manuscript evaluation, story beats, line edits, proofreading, consistency and AI detection.

Editorial analysis FAQ

Works with the rest of Familiar

The same manuscript feeds Familiar’s other editorial tools, and the book you publish.

Run it on your own manuscript

Create an account, add your manuscript, and run n-gram analysis for 1 credit.