Back to Glossary

Glossary

What is a Topic Cluster?

Updated September 2026ContentIntermediate

Definition

A topic cluster is a group of pages covering one subject from different angles, linked together around a broad page at the center. Unlike a keyword list, which treats every phrase as a separate target, a cluster treats the subject as the unit. Assistants split questions into parts, so coverage across those parts decides who appears.

At a glance

  • A cluster is a set of pages plus the links between them, not a category folder or a tag.
  • The center page covers the subject broadly. The others each answer one question properly.
  • Gaps in a cluster are invisible in rank reports and obvious the moment you ask an assistant.

Three pages beat one page three times as long

A question typed into an assistant rarely gets searched once. It is broken into smaller searches, each looking for a passage that handles one part of it, and the answer is written from what comes back.

So a subject covered by three focused pages can be retrieved three times, from three angles, for one user question. The same material as a single long page gets retrieved once, if at all, because its sections sit mixed together in fewer chunks.

That is the argument in one line. You are not trying to own a phrase. You are trying to be present in as many background searches as your subject generates.

Gaps in a cluster show up as somebody else’s brand

Rank tracking hides this problem completely. You can hold strong positions on every phrase you track and still be missing from answers, because the sub-question that triggered the retrieval was one you never wrote about.

A furniture retailer covering materials, dimensions and delivery, but nothing on how the frames hold up over a decade, loses the durability sub-search to a forum thread. Whoever wrote the durability page gets named in that part of the answer.

So audit a cluster from the answer side, not the keyword side. Ask the question, read what came back, and note which parts of your subject somebody else answered.

Map a cluster from the questions you already get

  1. Write the subject down as a plain phrase, the way a customer says it rather than the way your team categorizes it.
  2. List every question that falls under it. Sales calls and support tickets first, then the follow-up questions assistants suggest when you ask about the subject.
  3. Group them, then check which already have a page, which are buried inside another page, and which have nothing.
  4. Write the missing ones as their own pages, one question each, opened answer-first.
  5. Link every page to the center and to its closest neighbours, using the phrase already in the sentence as the anchor.

Do one subject completely before starting a second. A finished cluster on a narrow subject builds real topical authority. Three half-built ones build none.

A topic cluster next to a keyword list

  Keyword list Topic cluster
Unit of planning One phrase One subject
What you publish A page per phrase A page per question, linked
Sign of success Positions on tracked terms Presence across an answer’s parts
What a gap looks like A term you do not rank for A sub-question answered by someone else
Main failure Near-duplicate pages competing Clusters started and left unfinished

Frequently asked questions

  • Is a topic cluster the same as a pillar page?

    No. The pillar page is the broad page at the center. The cluster is that page plus every deeper page around it and the links joining them. A pillar page on its own is a long article with nowhere to send anyone.

  • How do I decide what belongs in a cluster?

    Group by the question, not by the keyword. If two pages answer the same question in different words, they belong together as one page. If a page answers a question your customers never ask, it does not belong in the cluster at all.

  • Can a cluster get too big?

    Yes, in two ways. Pages start duplicating each other, which splits your own coverage, and the set outgrows anyone’s capacity to keep it accurate. A cluster nobody maintains ages into a liability, because stale pages stay retrievable.

  • Do the internal links between pages really matter?

    More than people expect. They tell crawlers and models which pages belong to one subject and which is the entry point, and they carry a reader between related questions. Put them inside sentences where the subject comes up, not in a related-posts strip.