Keyword Clustering: How to Group Keywords Without Paid Tools

This guide teaches you keyword clustering with a manual method that needs only a browser and a spreadsheet, so you can decide which keywords share one page and which need their own.

What is keyword clustering and why does it matter?

Keyword clustering is the process of grouping related search terms so each group is targeted by one page. It matters because writing a separate page for every variation creates thin, overlapping pages that compete with each other, while forcing unrelated keywords onto one page leaves searchers unsatisfied.

Google’s SEO Starter Guide notes that people search for the same thing with different words and that Google’s language matching is sophisticated, so you do not need a page for every phrasing. Clustering turns that idea into a plan: one page per distinct need, with all the ways people describe that need covered naturally inside it.

How do you cluster keywords manually?

You cluster keywords manually by comparing their search results and grouping the ones that show mostly the same pages. Google already decided which pages satisfy each query, so overlapping results tell you the queries share an intent.

1. Build your list. Collect keywords from Search Console, search suggestions, People Also Ask and customer questions. Remove obvious duplicates.
2. Pick a head keyword. Start with the keyword that has the clearest intent or the most value to your business.
3. Record its top ten URLs. Search in a private window, set to your target country, and paste the organic URLs into a spreadsheet column.
4. Repeat for each candidate keyword. Record each keyword’s top ten in its own column.
5. Count shared URLs. For each keyword, count how many of its top ten also appear in the head keyword’s top ten.
6. Apply a threshold. Use the scoring table below to decide whether to group or split.
7. Name each cluster. The cluster’s main keyword becomes the page’s primary keyword. The rest become subheadings or natural variations.

Search results vary by location, language and device, as Google’s guide to how Search works explains, so always check from the market you are targeting. A UK and a US result page for the same keyword can differ enough to change a cluster.

The most common clustering mistake is grouping by wording instead of by results. “Carpet cleaning solution” and “carpet cleaning service” share three words but have completely different intents. Other common errors:

• Checking results from the wrong country or while logged in with heavy personal history.
• Ignoring SERP features such as maps, shopping results or videos, which signal a different page type.
• Re-clustering never. Results change, so recheck important clusters once or twice a year.
• Creating a new page for a keyword that an existing page already ranks for. Improve the existing page instead.

How many shared results mean keywords belong together?

As a working rule, keywords that share four or more of the same top ten URLs usually belong on one page, and keywords sharing zero to two usually need separate pages. This threshold is my practical rule, not a Google number, and borderline cases need judgement.

Look at page types as well as URLs. If one keyword shows guides and the other shows product pages, they need different pages even if a few URLs overlap.

For borderline scores, open the top three results for both keywords and read their headings. If the same outline would satisfy both searches, group them. If one search clearly wants a comparison and the other wants instructions, split them. Ten minutes of reading saves you from writing a page that half fits two intents and fully fits neither.

What is the difference between keyword clustering and topic clusters?

Keyword clustering decides which keywords go on one page, while topic clusters decide how several pages connect around a broader subject. Keyword clustering happens first and works at page level. Topic clusters work at site level.

A topic cluster is a pillar page plus supporting pages linked together. Each of those pages is the result of a keyword cluster. If you skip keyword clustering, your topic cluster often ends up with two supporting pages targeting the same need, which splits links and confuses readers. If you want a refresher on terms like pillar page or cannibalisation, my SEO glossary covers them.

How many keywords should be in a cluster?

A cluster should contain as many keywords as share the same intent, which can be two or thirty. There is no target size. Small clusters are normal for specific questions, and broad commercial terms often gather many variations.

What matters is that the page can answer every keyword in the cluster without becoming unfocused. If a page starts needing sections that would only interest one subgroup of searchers, that subgroup probably deserves its own page, linked from the main one.

A quick test I use: write the cluster’s main keyword as the H1, then list the other keywords as possible subheadings. If every subheading fits comfortably under that H1 and a reader searching any of them would be happy landing there, the cluster works. If one subheading feels like a detour, move that keyword out. In my content orders, clusters for informational guides usually end up with a handful of close variations plus several question style keywords that become H2s, while local service clusters stay small and tight.

How does manual keyword clustering work on a real list?

If you would like your keyword list clustered and turned into a content plan, that is part of my on-page SEO service. You can also check drafts against their cluster with the free SEO tools on this site.

Frequently asked questions