Topic Modeling
Statistical grouping of documents by the themes they contain, used to plan coverage.
Topic modelling is a statistical technique that groups documents by the themes running through them, without being told the themes in advance. In content planning it is used to see how competitors cluster a subject and which subtopics a site has never covered — the gaps that show up as an absent cluster rather than a missing keyword.
Definition
Classical approaches such as LDA infer latent topics from word co-occurrence; modern practice more often uses embeddings and clustering, which capture meaning rather than vocabulary overlap.
Applied to SEO, the input is usually the pages currently ranking for a set of queries. The output is a map of which themes those pages cover together — which is a better guide to what a comprehensive page needs than a keyword list.
Why It Matters
It moves planning from "which keywords do we lack" to "which parts of this subject do we not cover", which is closer to how retrieval systems judge topical authority.
Example
Clustering the top results for "core web vitals" reliably separates measurement, LCP causes, CLS causes and INP causes — four subtopics that a single overview page will not satisfy.
Related Terms
Know the term.
Check your own site.
A free audit tells you whether this is currently costing you score points — and exactly what to change.
Run a free audit