Keyword Clustering: How to Structure Content That Ranks

Keyword clustering is the practice of grouping search terms that Google answers with the same pages, so that one page on your site targets the whole group instead of each keyword getting its own thin article. You build a keyword list, check which keywords share results in the top 10, group the ones that overlap, and map each group to a single page. Done well, it decides how many pages you need, what each one should cover, and how they link together.

This guide covers what keyword clustering is, the two main ways to do it, a step by step process, and a worked example where we clustered the keywords for this very article using live search results.

Contents

• What is keyword clustering?
• Why keyword clustering matters
• SERP clustering vs semantic clustering
• How to do keyword clustering, step by step
• Worked example: clustering the keywords for this article
• Keyword clusters vs topic clusters
• Keyword clustering tools
• Common mistakes
• FAQ

What is keyword clustering?

Keyword clustering is a way to segment a list of target search terms into small groups, where each group belongs to one page of your website. The idea is simple: if Google already shows mostly the same pages for two queries, it treats them as the same need, and one well built page can rank for both.

The method most tools use today looks at the first ten search results for each keyword. When two keywords return enough of the same URLs, they go into the same cluster. Wikipedia's article on keyword clustering credits the term and its first principles to the Russian SEO specialist Alexey Chekushin in 2015, along with the first automated clustering tool built on search results.

Before that, SEOs mostly grouped keywords by shared words (often called lemma based grouping): "keyword cluster", "keyword clusters" and "clustering keywords" went together because they share a root word. That catches spelling variants, but it cannot tell you whether two differently worded queries want the same page, or whether two similar looking queries want different ones.

Why keyword clustering matters

• One page can rank for dozens of queries. Pages rarely rank for a single keyword. Clustering makes that deliberate: you write one complete page for the whole group instead of hoping it picks up the variants.
• You avoid keyword cannibalization. When two of your pages target the same intent, they compete with each other and Google often rotates between them. A cluster map shows you, before you publish, that the second page should not exist.
• You publish fewer, stronger pages. Instead of 40 near duplicate posts, you might need 8 pages, each covering one need properly.
• It builds topical authority. Clusters show which subtopics a subject is made of, which is the raw material for a pillar and supporting pages structure. See our glossary entry on topical authority.
• It gives every page a clear brief. The cluster tells the writer which secondary terms and questions the page has to answer.

SERP clustering vs semantic clustering

There are two main ways to decide which keywords belong together.

SERP based clustering

SERP clustering compares the actual search results. For each keyword, a tool collects the top 10 results, then counts how many URLs each pair of keywords shares. If the count reaches a threshold you set, the keywords are grouped. Wikipedia calls this threshold the clustering level: a higher level produces more groups with fewer keywords in each, a lower level produces fewer, larger groups.

Tools also differ in how strict a group has to be. In a soft cluster, every keyword shares results with the main (highest volume) keyword, but not necessarily with each other. In a hard cluster, every keyword shares results with every other keyword in the group. Hard clusters are tighter and safer to put on one page; soft clusters are broader and better for planning.

• Pros: it reflects how Google currently understands intent, it works across languages, and it catches pairs no word matching would find.
• Cons: it needs live search data for every keyword, which costs money at scale, and results shift by country, device and over time.

Semantic clustering

Semantic clustering groups keywords by meaning, using word overlap, embeddings or a language model, without looking at search results.

• Pros: fast and cheap, works on huge lists, and useful for a first pass or for brainstorming topics.
• Cons: it groups by what words mean, not by what Google ranks. Two queries can mean nearly the same thing and still need different pages, as the example below shows.

A practical approach is to use semantic grouping to organize a raw list into themes, then confirm the groups that decide page structure with SERP data.

How to do keyword clustering, step by step

1. Build and clean your keyword list

Start from your keyword research tool, Google Search Console queries your site already gets, competitor keyword exports and the "People also ask" questions for your main terms. Remove duplicates, misspellings you do not want to target, and anything off topic. Keep search volume next to each keyword; you will need it to name clusters.

2. Collect the top 10 results for each keyword

Use the same country, language and device for every keyword, and collect them in the same session if you can, because results move. Record only the organic URLs. Rows such as AI Overviews, People also ask and video carousels are not URLs you can compare.

3. Compare overlap and choose a threshold

For every pair of keywords, count the shared URLs. Pick a threshold before you look at the groups, so you are not tuning it to the answer you want. A higher threshold keeps clusters tight; start strict and loosen it only if you end up with too many single keyword groups.

4. Name each cluster by its head term

The keyword with the highest search volume usually becomes the cluster's primary keyword, which goes in the title, H1 and URL. The rest become secondary terms and subheadings.

5. Map clusters to pages

For each cluster, check whether a page on your site already ranks for any of its keywords (Search Console's page report shows this). If it does, improve that page rather than writing a new one. If two of your pages split one cluster, merge them and redirect. Only create a new page when no existing page fits.

6. Structure the pages and link them

Group related clusters under a pillar page and link the supporting pages to it and to each other where a reader would naturally go next. Our internal linking analyzer helps check those links, and the content brief generator turns a cluster into an outline.

7. Track rankings and re-cluster

Track the whole cluster, not just the primary keyword. Search results change, so re-run the clustering for important topics every few months and after major Google updates. A cluster that splits apart is a sign the intent has changed.

Worked example: clustering the keywords for this article

Before writing this post we ran the process on its own candidate keywords. We pulled the top 10 Google results (United States, English, desktop) for six related queries on October 15, 2026, through OpenSEO, and counted the organic URLs each query shares with "keyword clustering". Monthly search volumes are United States estimates from the same source.

Query	US monthly searches	Organic URLs shared with "keyword clustering"
keyword clustering	480	(the reference query, 8 organic URLs)
seo keyword clustering	70	5
keyword clustering tool	720	3
best keyword clustering tool	110	1
what is keyword clustering	110	0
keyword grouping	70	0

What the overlap says:

• "keyword clustering" and "seo keyword clustering" are one cluster. They share 5 of 8 organic results. One guide should target both, which is what this page does.
• The tool queries form their own cluster. "keyword clustering tool" and "best keyword clustering tool" share 5 results with each other, mostly tool pages and "best tools" lists. Searchers there want to compare products, so a how to guide is the wrong page for them, even though "keyword clustering tool" has the most searches of the set.
• "what is keyword clustering" shares nothing with "keyword clustering". Any semantic tool would put these two together, since one is literally a question about the other. Google showed a completely different set of pages: explainers, a case study, a Chrome extension and a Facebook answer page. That is exactly the kind of split SERP clustering catches and word based grouping misses.
• "keyword grouping" is a separate need too, despite being a synonym. Its results were a forum thread, several grouping tools and tool roundups.

One more observation: every one of the six queries showed an AI Overview above the organic results. For informational SEO topics, being the page that answers the question clearly in its first lines matters as much as ranking position, which is why this post opens with a direct definition.

If you want to run the same comparison yourself on a small list, a few lines of Python are enough once you have the top 10 URLs for each keyword:

from itertools import combinations

serps = {
    "keyword clustering": [...],      # top 10 organic URLs
    "seo keyword clustering": [...],
}

for a, b in combinations(serps, 2):
    shared = len(set(serps[a]) & set(serps[b]))
    print(f"{shared}  {a}  <->  {b}")

Keyword clusters vs topic clusters

The terms are often used interchangeably, but they work at different levels.

• A keyword cluster is a group of search terms that one page should target. It answers "which keywords go on this page?"
• A topic cluster is a group of pages: a pillar page covering a broad subject, plus supporting pages that each cover one subtopic in depth, all linked together. It answers "which pages does this subject need?"

Keyword clustering is how you build topic clusters on evidence instead of guesswork. Each keyword cluster becomes one supporting page (or the pillar), and the overlaps between clusters tell you which pages should link to each other. Our glossary explains content clusters, and we compare content clusters vs individual posts if you are deciding between the two approaches. For the technical side of making those pages crawlable and fast, see our technical SEO guide.

Keyword clustering tools

Most tools fall into two groups:

• SERP based clustering tools fetch live results for every keyword and group by shared URLs. Keyword Insights, KeyClusters and ContentGecko's free clustering tool describe themselves this way, and Semrush says its Keyword Strategy Builder uses SERP based clustering too. Because every keyword needs a live search, these tools usually cap or charge by the number of keywords.
• AI or semantic grouping tools, including several free ones, group keywords by meaning. They are quick for brainstorming and first passes, but check any cluster that will decide your site structure against real search results.

For small lists, a spreadsheet works: paste the top 10 URLs for each keyword into columns and count matches. It is slow, but it teaches you how the method works, which makes any tool's output easier to judge.

Common mistakes

• Clustering by words alone. As the example shows, similar wording does not mean similar intent.
• Mixing markets. Results for the United States and the United Kingdom, or desktop and mobile, can differ. Cluster each market separately.
• Writing one page per keyword anyway. The point of clustering is fewer, better pages. If two keywords share most of their results, they belong on one page.
• Ignoring pages you already have. Check what already ranks before you create anything new.
• Treating clusters as permanent. Search intent drifts; re-check important clusters regularly.

FAQ

What is keyword clustering in SEO?

It is grouping keywords that share search intent, usually measured by how many of the same pages Google shows for them, so that one page targets the whole group.

How do you cluster keywords?

Collect your keyword list, get the top 10 results for each keyword, count shared URLs between every pair, group the pairs above your threshold, and assign each group to one page.

How many shared results should put two keywords in the same cluster?

There is no single right number; most tools let you set it. A higher threshold gives tighter clusters that are safer to put on one page. Start strict and loosen it only if you get too many single keyword groups.

What is the difference between a keyword cluster and a topic cluster?

A keyword cluster is a set of keywords for one page. A topic cluster is a set of linked pages, a pillar plus supporting articles, that together cover a subject.

Is keyword clustering still worth it with AI search?

Yes. AI Overviews appeared on every query in our example. Clustering makes each page answer one need completely, which is what both search rankings and AI generated answers draw on.

Can I do keyword clustering for free?

For a short list, yes: compare the top 10 results by hand in a spreadsheet, or use a free AI grouping tool for a rough first pass and check the important groups against live results.