Lexical Diversity Calculator

Lexical diversity is the ratio of distinct words to total words, and it exists to answer a question raw vocabulary counts cannot: how much of this text is new information versus restatement. A type-token ratio of 0.5 means half the words are first appearances. Conversational speech typically runs low, edited prose sits in the middle, and technical reference material runs lower still because it repeats defined terms deliberately.

The metric's well-known weakness is length sensitivity. Every text reuses common function words, so the ratio falls automatically as a document grows, and a 200-word passage will always score higher than a 2,000-word article by the same author on the same topic. This makes cross-length comparison meaningless. Compare passages of similar length, or compare successive drafts of the same piece, where length is roughly held constant and a change actually signals something.

High diversity is not a quality target, and chasing it produces worse writing. Reaching for a synonym so as not to repeat a term is exactly the wrong instinct in documentation, where a reader needs to know that two mentions refer to one thing. The genuinely useful reading is a low score in writing that should be varied — a landing page, a narrative section — which usually means a few phrases are being cycled and the prose has started to feel like padding.