Our partnerWriter MarginsKDP profit trackingfree plan Beta: 50% off Pro
← Back to resources

What Term Clustering Tells You

An invented word stops registering as unusual to the person who invented it, so its mentions can pile up in one stretch of the book without the writer noticing. The reader meets it four times in a chapter and then not again for a hundred pages. Clustering shows where that happens, and it turns out to answer a second question as well: whether a term is a working part of the book or a prop for one scene.

The two questions it answers

QuestionWhat the report shows
Is the reader being asked too much of a term at once?Mentions bunching close together in the text
Does the term earn its place?Whether its use spreads through the book or sits in one scene

The first is a pacing problem, and it concentrates in openings and in the scenes where a new part of the world is introduced. The second is a world-building problem, and it's the one authors find more surprising.

A cluster of one

A coinage whose mentions all sit in one scene is an idea that was named and not built. Sometimes that's right: a single word can imply a whole institution. More often it is a thing you invented for one scene and never returned to, and the reader experiences it as a word that meant nothing.

The decision is to develop it or cut it. Developing means giving it two or three related terms so it reads as part of a system; cutting means using an ordinary word instead. Both are better than leaving one unexplained coinage standing alone.

A crowded page

Four unfamiliar words in one paragraph is a paragraph a reader decodes rather than reads, and openings are where this happens because openings are where writers front-load. The remedy is almost never to explain more; it's to introduce fewer, later. A term first met in chapter six is met by a reader who has capacity for it.

The related check is the first appearance index, which counts how many things enter in each chapter. A first chapter introducing twenty-five things is the diagnosis for a first chapter readers describe as hard to get into.

Running it

The Novel Kit's term clustering report shows where each term's mentions bunch in the manuscript, measured against that term's own average rate. Read it alongside two others: world-building density, which measures where the invented vocabulary sits across the whole book, and unexplained terms, which finds the ones the book leans on without ever defining.

A cluster of unexplained terms in one early scene is the worst combination and the most common: several new words, none defined, in the place where a reader is deciding whether to continue.

No report knows whether a term deserves to exist. It tells you where the words fall and how they bunch, and the decision about which to keep is a craft one.

A practical next step

Look at your clusters and find the terms whose mentions never leave one scene. Each is a decision: develop it, or replace it with an ordinary word. Most authors find two or three, and cutting them makes the remaining vocabulary work harder.

Sources checked

Frequently asked questions

What is a term cluster?

Invented words that arrive close together in the text. Grouping them shows where the world-building actually lands rather than where you think it does.

Why does clustering matter to a reader?

Because a reader meeting four unfamiliar words on one page is decoding rather than reading. Spread across four pages the same words are effortless.

What does a single-term cluster mean?

Usually an idea that was named and never built. Worth either developing or cutting, and it is the finding authors act on most.

Related reading

See how your terms group

One pass lists every name, place, faction and invented term in the book.

Map your manuscript