keyword clustering

Since April 2023, we’ve managed to boost SEOTesting’s organic traffic from approximately 2,000 visits per month, to over 15,000 visits per month.

How did we do this? By using keyword clustering when writing our blog content.

In this blog – we’ll explain what keyword clustering is and how it can benefit your SEO performance. We’ll then uncover some different keyword clustering methods you can try out for yourself. We’ll also talk about how you can cluster your keywords manually and how you can cluster keywords using tools. You’ll then learn how to use URL switch tests to check whether keyword clusters are working and how content groups can be used to monitor their performance over time.

What is keyword clustering?

Keyword clustering is a term used to describe a group of keywords/queries that a single piece of content ranks for. This process removes the need for marketers and copywriters to publish one single piece of content for each keyword they want to target, saving:

Instead, you can conduct keyword research and then categorize your keywords into relevant groups known as clusters. This means you only need one piece of content to target each cluster.

Here’s an example of how this looks on a website:

The content is focused on one particular keyword, in the example above it is keyword research guide. From there, there are several other keywords included in the cluster including examples like:

Why does keyword clustering work?

Keyword clustering works because it utilizes the relationships between keywords to create organized, related groups of content. This method leverages the way search engines now process and rank content that is topically relevant and comprehensive, rather than solely focusing on keyword matches.

Here’s why keyword clustering is so effective:

Improved topical relevance: By grouping related keywords, you can create content that addresses a broader spectrum of a topic. This helps to enhance the topical relevance of your content, and makes it more likely to rank well for a variety of related search queries.

Better user experience: When content is structured around keyword clusters, it offers users a more coherent and valuable experience. They can find answers to multiple questions in one place, which reduces their need to search elsewhere and land on competitor websites.

Optimized for search intent: When you put together keyword clusters, you can better understand and cater to different user intents within the same topic. Covering various aspects of a topic can satisfy informational, navigational, and transactional search intents, which is essential to ensure your content ranks across different stages of the user journey.

Content creation efficiency: Rather than the “old” process of creating one piece of content for each keyword, clustering allows you to cover multiple keywords with fewer pages. This saves you time and money and helps build a more authoritative site structure.

Internal linking opportunities: Keyword clusters naturally make it easier for you to create a strong internal link structure. Linking between related content strengthens your site’s topical relevance and helps search engines understand the context and importance of different pages.

Keyword clustering examples

There are two different types of keyword clustering – semantic keyword clustering and SERP keyword clustering.

Below, we’ll explain these two approaches to grouping your keywords, to help you decide which method is best for your business.

Semantic keyword clustering

Semantic keyword clustering is a technique for grouping related keywords based on their meanings and context, rather than just matching words that look similar.

Let’s say, for example, that you’re creating content for a fitness website. Instead of targeting just one keyword like “exercise routines,” you could use semantic keyword clustering to group related keywords together, such as:

These keywords all relate to the broader topic of exercise routines, but each addresses a specific area of the topic. Creating a single, comprehensive article (or a small series of articles) that covers all of these related topics can attract more visitors searching for related information.

Here’s how this would look in action:

SERP keyword clustering

SERP keyword clustering is a method of clustering whereby keywords are grouped together based on how they appear on search engine results pages.

If different keywords lead the user to similar search results, it likely means that the same page can be used to target these multiple keywords. By clustering these keywords, you can create content that targets multiple search terms at once, improving your chances of ranking higher for a range of related queries.

Let’s say you worked in marketing for a travel company, and wanted to create content for tourist attractions in Paris. It’s likely that you can make one piece of content to target several keywords in that sector.

The above screenshot shows a SERP similarity of 73.68% between the following keywords:

This means there is a good chance you could create one piece of content to target both of these keywords.

How to create keyword clusters manually

Before we explain how to create keyword clusters using tools, we’ll show you how to do it manually and at a low cost.

As long as you have access to spreadsheet software such as Microsoft Excel or Google Sheets, ChatGPT, and a rank tracker, you can follow the steps below:

Step 1: Build a list of keywords

First, you’ll need to build a list of keywords/queries you want to target with your content. This list of keywords will come from different sources, including:

You’ll need to build a detailed list of both long-tail and short-tail keywords that cover different intents to ensure you create content for all users at every stage of the funnel.

Looking for a good place to start? You can use Ahrefs’ Keyword Explorer and type in a head keyword. For example, I typed in running shoes and searched for them in the US. Next, head to the Matching Terms section, and Ahrefs will provide you with a massive list of keywords to start you off.

You can export this to start your keyword list on the right track, giving you a spreadsheet of different keywords with different search intents:

Use all the keyword research resources listed above to find other related keywords, add them to your spreadsheet, de-duplicate them, and you’ll have an exhaustive list of keywords around the topic to work with.

Step 2: Categorize your keywords by search intent

Your next step is to categorize your keywords by their search intent. All keywords/queries that you target will have one of four possible intents:

Identifying the intent of your keywords is important because it will inform the type of content you need to create. Informational blog posts won’t work for transactional queries, will they? So, you need to make sure that these are separated so your content has the best chance of success.

Luckily, ChatGPT can help speed this process up for you:

As you can see above, ChatGPT has provided a table that includes the keywords and their intent, which can be downloaded. To achieve this, we used the below prompt:

“I am giving you a CSV file which contains [number of keywords] for you. Please classify each keyword into four categories: Informational, Navigational, Commercial, and Transactional. Return the keywords and their intents in a table.”

We can then download this table of keywords with their search intents already mapped out.

Step 3: Cluster your keywords

The next step is to cluster your keywords so you know what content you need to create to target each group. As we mentioned earlier, there are two ways to cluster your keywords. You can cluster them semantically, or you can cluster by SERP similarity.

Clustering keywords semantically

The goal of clustering your keywords by semantic similarity is to group together keywords that convey a similar meaning. This allows you to create content that comprehensively covers each cluster, maximising your chances of gaining clicks and driving business impact with the smallest number of content pieces.

This process can be time-consuming, as it requires analysing the semantic relationships between your keywords. Luckily, you can use ChatGPT to help you with this. You can use ChatGPT to efficiently group keywords together that share similar meanings.

Here’s an example of an effective prompt:

“I am uploading a table with 250 keywords. I want you to cluster these keywords based on their semantic similarity and provide the results in a digestible table. Ensure that keywords with similar meanings are grouped together.”

This prompt ensures that ChatGPT understands:

ChatGPT will then deliver precisely what you need:

With these semantic clusters, you now have groups of keywords that share similar meanings, which you can target with your content. For example, in the provided screenshot, the following keywords have been clustered together:

Given that all these keywords relate to how running shoes fit, you can create a comprehensive guide titled “Finding the Best Running Shoe Fit,” with dedicated sections on overpronation, supination and overall support.

Clustering by SERP similarity

You can also choose to cluster your keywords by SERP similarity. This method may work best if you are looking to publish a lot of content and organic traffic is a significant growth lever for your business.

Keep in mind that by clustering by SERP similarity, there is a chance you’ll end up with clusters that include keywords of differing intents. So, you must find a way to balance this in your content creation and optimization.

Clustering keywords by SERP similarity involves looking at the number of similar results that appear in the SERPs for different keywords, as we explained earlier. The trouble with this is that it can take a lot of time and effort to do manually.

You must search Google for each keyword and note the top 10 URLs returned. You’d then need to manually place each keyword into groups based on the level of similarity.

Keyword Insights, for example, includes keywords in a cluster if four or more URLs are similar.

Once this is done, you’ll have a list of keyword clusters to start targeting.

You can also use ChatGPT to help you create your clusters, but you’ll need to do the manual work first (or use a SERP scraper) to get your list of keywords and the URLs returned for each keyword.

Step 4: Create/optimize your content

Once you have finished clustering your content, either by SERP similarity or semantically, your next step is to create content for each cluster or optimize your existing content already available on your site.

For each new piece of content created or piece of content optimized, make sure you include the keywords in your cluster within your content to create relevance. Above all, you should create content that is valuable to the user.

Step 5: Track keyword rankings

Finally, you’ll want to find a way to track the success of your content clusters, and one of the best ways to do this is by using a rank tracker. SEOTesting has a free rank tracker available that allows you to track up to 500 keywords, so if all of your keywords total 500 or less, you can monitor these for free:

How to do keyword clustering using tools

Luckily, there are several different keyword clustering tools out there that will help you do the job quicker. Let’s use the same example of keywords surrounding running shoes, and cluster these (based on SERP similarity) using Keyword Insights.

We’ll first start by setting up our clustering project, as seen below:

Here, we let Keyword Insights know that the project name will be ‘Running Shoes,’ and that we want to cluster based on English-language SERPs in the United States. We don’t need the rankings for this particular project, but we do want Keyword Insights to classify the search intent of the keywords we import.

We’ll then click next so we can import our keywords.

We’ve uploaded our CSV of keywords and selected the relevant keyword and search volume rows. Now, all that is left to do is wait for Keyword Insights to analyze and cluster the keywords for us.

According to the software, this should take about 5 minutes:

Once this is done, we can export the keyword cluster document to Google Drive and see the content we need to create based on the keyword clustering from Keyword Insights!

Thanks to Keyword Insights, it took less than 10 minutes to create a keyword cluster document that includes 500 keywords about running shoes. We also have information on:

All that’s left to do now is create user-first content that targets your keywords.

Testing keyword clusters using URL switch tests

One of the best ways to test content clusters is to utilize URL switch tests.

Using URL switch tests, you can establish whether your content performs better as one long-form piece of content targeting multiple keywords, or as smaller, individual articles targeting different search terms.

Let’s take a look at this test as an example:

The above screenshot shows the results of a URL switch test we ran to decide whether having one large or multiple smaller pieces of content was most effective. We were testing whether targeting a cluster of keywords for this particular group of queries was worth it.

We had one long piece of content on keyword cannibalization:

Keyword Cannibalization – how to find issues, and how to fix them

What is Keyword Cannibalization? - How to find issues and fix them

SEOTesting has a report that helps find keyword cannibalization issues across your site that affect your top keywords.

In an ideal world each page on your website would target a specific keyword, but as websites age and more content gets published – it can be harder for Google to decide which page of your site is best to rank for a specific query. The keyword cannibalization report is really useful to run as part of a full site content audit.

YouTube

You can find the keyword cannibalization report by selecting the site you wish to work with and visiting the Reports section from the left hand menu.

Upon clicking the button to request the cannibalization report, SEOTesting will start processing your site’s data and send you an email when the report is ready to view.

This report takes the top 20% of queries that your site appears for in Google, and checks whether each keyword has multiple pages ranking.

If a query has 2 or more pages from your site being displayed in the search results this could indicate 2 or more of your pages are competing against each other. Fixing this issue could really improve your ranking for that keyword.

What is Keyword Cannibalization in SEO?

Keyword cannibalization, when referred to in the context of SEO, essentially means that multiple pages from the same website are targeting the same (or very similar) keywords in Google.

In Layman’s terms, two pages from the same website come up on the same Google search engine results page.

Here’s an example for you to use. Let’s say we have written two guides on “SEO Reporting” which have very similar titles and very similar, even if it isn’t identical content. You may find that both of these pages start ranking for the term “SEO Reporting” on Google.

Is Keyword Cannibalization Bad for SEO?

On the face of it, keyword cannibalization seems like it could be a really good thing! “Great, we’re getting more of our web pages on Google’s SERPs. This is really good.” Unfortunately, that isn’t quite the case.

If you have two posts on the exact same topic, you will actually find these two pieces of content competing for spots on the search engine results page, harming your organic performance. Both of these pieces will find themselves lower down in the SERP when you could be much higher with a combined effort. We’ll talk more about that a little later in the article.

Sorry to have to be the bringer of bad news, but having multiple pages included in the same SERP can actually hurt your organic search performance.

Instead of taking up more SERP real estate and dominating Google, you will (in most instances) find that Google doesn’t quite understand which page to rank highly, and marks both of them down as a result.

It’s not a penalty from Google, so please don’t start panicking! But Google could certainly be confused as to which page to rank higher and rank both pages lower in the search results, limiting organic performance.

Why Is Keyword Cannibalization Bad for SEO?

So we have explained above that, in most cases, it is bad to have instances of keyword cannibalization across your website. There are a few reasons for this…

If you have lots of instances of keyword cannibalization, you are going to be squandering a lot of your crawl budget. This is not an issue for most sites, but for large eCommerce websites, news sites and other content aggregators then it could be a big issue that needs fixing right away.

You may find that your conversion rate suffers. No page on any website is going to have exactly the same conversion rate as another page, one will absolutely convert visitors much better than the rest. Instead of allowing visitors to choose which page to go to, why not redirect the lesser converting page into the more converting page and make more money?

Google may even devalue a more relevant page! If you have a lot of keyword cannibalization going on around your website, Google will inevitably try to understand which page is the best fit for that particular set of search results. Google does get it wrong sometimes. For example, if you have two similar pages ranking for the same keyword, Google may decide to rank the page that converts more visitors into leads lower than the other page, leaving a lot of money on the table for your business.

Any Examples of Keyword Cannibalization Being Good for SEO?

In general, you should only aim to have one page available to serve for each keyword and their variations. This will ensure that page is given the most authority from Google and other search engines like Bing and Yahoo and will give that particular page the best chance of ranking higher, especially as you build more relevant and trustworthy backlinks to the page!

If you are, however, really wanting to dominate the SERP with your website as much as possible, it is worth researching and seeing if there is an opportunity to implement the Surround Sound SEO Strategy into your search engine marketing.

How to Find Keyword Cannibalization Issues (The Hard Way)

If you are not a user of SEOTesting, there are a few different things you can do to find keyword cannibalization opportunities on your website.

  1. You can run a content audit, listing all of the pages of your website and noting down key pieces of information like the content titles, meta information and key information included in the content. This will help you establish which keywords are being targeted by which page. This only works (most of the time) with small to medium websites.
  2. You can run a site search in Google using the “site:” search operator. Simply search all instances where Google has indexed and ranked your website and add different topics to find multiple pages that are serving the same keyword or topic.

For example, if I was to try and find instances where our website was ranking for two different pages with the “SEO Reporting” topic, I’d simply write this into Google: site:seotesting.com “SEO reporting” and this would list all instances of our website that are ranking with the words SEO Reporting somewhere in the title.

How to Find Keyword Cannibalization Issues (The Easy Way)

As I mentioned at the very start of this guide, SEOTesting provides a really simple way for you to find all of the pages on your website that rank for the same keyword.

Simply head to your SEOTesting dashboard, click “Reports” and then click “Keyword Cannibalization” which will be right at the very bottom of your reports page. Click “Request Report” and in a short period of time, your report will be emailed to you!

You’ll also be able to see a list of past reports located in the section as well. This is incredibly handy as you’ll be able to compare recent reports to reports you have just created, allowing you to monitor how well you are doing at fixing issues of keyword cannibalization.

How to Fix Keyword Cannibalization Issues

Check your Page Titles

You may have a simple issue of duplicate page titles, this is incredibly common on eCommerce websites where products may be listed within different categories.

If so you may be able to remove the target keyword from any page that is not the target to rank for the keyword.

As an example. If I have an eCommerce website and I have men’s jeans in a wide array of categories, like 32” waist jeans and 34” waist jeans, these titles, even though they serve completely different products, may be duplicated. This could cause an array of cannibalization issues across your site. As a fix for this, you could rename your pages so they are only relevant to the size they are serving.

Link from the Non-Target Pages to the Main Page you want to Rank for that Keyword

If you do have multiple pages ranking for a keyword, and you want to keep all the pages, you should internally link from non-target pages to the target url. This link should use the keyword as the anchor text as this will indicate to Google that this target url is the one that should be ranking for this keyword.

Locating keyword cannibalization issues can be turned into a positive for a site by showing you lots of opportunities to include additional internal links across the site.

Merge Pages and Redirect

There may be the possibility of taking multiple pages ranking for the same keyword and combining them into a single comprehensive page.

If you do this, make sure you redirect any urls you combine to the url you are targeting for the keyword.

A good example for merging pages and redirecting them would be the example we have used throughout this particular guide. SEO Reporting. If we were to have written two guides on SEO Reporting with very similar content, there is almost always the opportunity to include some of the content from one of the guides and add it into the other guide. We can then redirect one of the guides to the other page.

This, in most cases, will drive additional authority to Google and allow the guide that is not redirected to rank a little higher, driving more impressions, more clicks and more leads!

Remove Pages and Redirect

Rather than merge articles together, it may be a simple task of removing unnecessary pages that accidently target a keyword. As with merging, make sure you redirect any urls you remove to the page you do want to rank for the keyword.

Do Nothing

Sometimes having 2 results in the search results for a keyword is what you want, especially if they fulfill different types of search intent for example an ecommerce page and a review article.

Frequently Asked Questions

What is keyword cannibalization?

Keyword cannibalization happens when multiple pages on your site target the same or very similar keywords. This can confuse search engines and hurt rankings.

Why is keyword cannibalization bad for SEO?

It splits ranking signals between pages. Google may rank both lower. It can also waste crawl budget and hurt conversions if the wrong page ranks.

How can I find keyword cannibalization issues?

You can use SEOTesting’s Keyword Cannibalization report. It shows keywords where two or more pages on your site rank. You can also run a site search or do a manual content audit.

How do I fix keyword cannibalization?

You can merge pages, redirect duplicates, adjust titles, or link to the main target page. Internal linking with the keyword as anchor text helps signal the right page to rank.

Is keyword cannibalization ever good for SEO?

In rare cases, yes. If two pages serve different search intents, having both appear can help cover more user needs. But this should be intentional.

Summary

Keyword cannibalization could be harming your efforts to rank specific keywords. I recommend running the Keyword Cannibalization report at least once a quarter. This will help you diagnose cannibalization issues so you can fix them and give your target page the best chance of ranking well.