Group Keywords by Relevance

The SEO Keyword Research Master Guide - Chapter 4

Group related keywords

Last Updated: March 13, 2023

Keyword Research isn't only about finding the right keywords, it's also about finding the right related or semantic keywords to group together.

Related keywords are important because they help build your content and add topical relevance, helping your website to rank.

Click to Tweet

Semantic keywords can also help you to rank for a broader set of terms. Often, you'll find a popular, well-optimized page ranking for hundreds or thousands of related keyword phrases.

Group keywords

When creating website content, you're typically faced with two keyword challenges:

  • When to create separate pages for each keyword vs. grouping them together

  • Which related keywords should you include with each topic?

We'll address these each in turn.

When to create separate pages for each keyword

Let's consider our keyword suggestions for "motorcycle jackets".

Is it worth it for us to create a separate page for each of these keywords? Probably not, but how do we decide?

Here are a few criteria to consider when creating a unique page for a keyword.

Grouping keywords

1. Sufficient volume

Does the keyword get enough search traffic to warrant its own page?

In the example above, "motorcycle jackets amazon" records only one search per month. For many people, this isn't worth the effort.

(Also, you don't want to have to compete against Amazon for one lousy visit per month. Your time can be better spent on less competitive, lucrative keywords.)

2. Is it a separate topic?

Some related keywords are so similar to each other, they obviously talk about the same thing. In the old days of SEO, you could get away with splitting each and every related topic variation onto its own page, and Google could rank you.

Since Google released its Hummingbird algorithm (and later, Rankbrain), it's gotten much smarter at grouping keyword intent and figuring out when separate keyword phrases have the same intent. Natural Language Processing - NLP - has allowed search engines to understand natural human language better than ever before. Likewise with the launch of the Google Knowledge Graph, SEOs now group keywords by topic or entity.

Google defines entities as ‘a thing or concept that is singular, unique, well-defined and distinguishable'. Modern SEO uses entities to dig deeper, and entity-based SEO shifts the focus to entities and how they relate to each other, rather than just singular keywords.

For example, in the above keyword set we see the following phrases:

  • motorcycle jackets leather: 47

  • leather motorcycle jackets: 677

These seem obviously like the same topic, but often it's easy to misjudge intent. Someone could be looking for leather to make a motorcycle jacket, or they could be looking to buy a leather motorcycle jacket. To be sure, you can simply Google the two keywords and see how much overlap there is between ranking results.

If both keywords return the same general set of URLs, this is an indication that Google considers these two keywords synonymous enough that you can group them together - they are semantically connected keywords.

Separate topic keywords

In this example, both keywords returned the exact same URLs, so we know we don't need to split them out onto their own pages.

3. Does it make sense?

Presumably, you know your subject matter better than most.

When creating content, you should ask yourself if a page deserves its own topic, or if it should be combined with others.

You don't have to follow the crowd. Google results are a great indication of what people are searching for, but it's fine to forge your own path if you have something new or important to offer. After all, creating content that is unique can help your page rank higher.

Joelle Irvine: Grouping Keywords
Joelle Irvine: Grouping Keywords

"Beyond ranking for more keywords, broadening the focus of a page with semantic language and related topics can help your audience find what they’re looking for faster. To maximize success on-page, leverage related keywords in subheads to support skimmers as they scroll for answers, pique the interest of those doing research, and encourage all to read more! These optimizations also help search engines’ machine learning capabilities, notably Google's Natural Language Processing, provide more relevant results by matching query context and intent with the content on your page. To take it a step further, use schema markup to connect your site's entities with those in the wider knowledge graph and – if all the stars align – potentially gain some highly-coveted featured snippets."

Joelle Irvine - Director of Growth Marketing, Moz

How to group keywords together

When planning content, you obviously start with your main keyword, but it helps to have a list of 5-20 other related keywords to help build out your content.

These closely related keywords help inform what the content is about, can become your subject headings, and generally provide additional structure and meaning to your content.

Learn how to explore hundreds of keywords and create your very own keyword list with the Moz Academy Keyword Research Certification.

Out of a list of potentially hundreds of semantic keywords, how do you choose the most important, most relevant phrases?

1. Relevance

First and foremost, your chosen related keywords must be relevant to your main focus keyword.

Relevance is often the easiest to figure out because we've already done all the work in the steps above.

To make it easy, Keyword Explorer actually contains a relevance metric when you research keyword suggestions. When we've narrowed down our main topics, we can simply sort each suggestion by relevance.

The score is a little easier to work with if you export the data into a spreadsheet, as in this example.

Keyword relevance

By this point, you should have a pretty good understanding of which keywords are most related to each other.

2. Volume

We keep coming back to volume because it's such a valuable metric.

When deciding which related semantic keywords to use, you can use volume metrics to filter out lower-value keyword opportunities.

In the example above, "motorcycle jackets with armor" has the following search volume metrics:

  • Minimum volume: 501

  • Maximum volume: 805

  • Exact volume: 730

Minimum and maximum volume represent the estimated range of monthly keyword search volume with the highest probability, whereas exact volume represents a best guess estimation of the exact search volume for that keyword.

If it fits the brand or offering, we'd definitely want to consider "cheap" in this context.

On the other hand, "mesh motorcycle jacket with armor" only has a search volume of 1-10. If "mesh" wasn't central to our offerings, we'd probably consider leaving this keyword out of consideration.

3. Modifiers, n-grams, and repeating phrases

When you see the same keywords repeating over and over again in your keyword set, that's a good sign you should include those topics and phrases in your related keyword set.

This is true even if the individual keywords don't have very much volume by themselves, because as we know, having a bunch of long-tail keywords in our content can eventually add up, allowing for more gradual traffic as well as attracting new individuals to our content.

If we look at our "motorcycle jackets with armor" example, we can see the same keyword modifiers appearing again and again. These include:

  • cheap

  • mens

  • leather

These appear so often with our main keywords, it seems like a good idea to include these modifiers when crafting our on-page keyword strategy, assuming that these keywords are relevant to the page.

Often, it's helpful to use a tool such as a word frequency counter to actually count all your semantic keyword variations. We've loaded the first 50 most relevant keywords from our spreadsheet into such a tool and came up with the following word frequencies:

Keyword Frequency
motorcycle 47
armor/armored 33
jacket/jackets 32
with 11
body 6
best 6
leather 5
for 4

In this case, "body", "best", and "leather" were a few of the most frequent keywords to appear, so we might consider looking at phrases that include these terms.

Another helpful technique is to find the n-grams in your keyword set. N-grams are simply sequences of words of a certain number.

You can use an n-gram viewer, such as this one by SEO Ryan Jones, to help visualize the most important n-grams in your keyword set.

Keyword n-grams

Here we can see common patterns start to emerge, including:

  • cheap

  • body armor

  • gear

  • protective

It's important to remember when building your related keyword phrase list, that you may not use these terms exactly as they appear here - remember entities.

Instead, you're looking for broad patterns and modifiers that you can use to build a list of a few relevant, related keywords to support your primary keyword.

Pro tip: finding related keywords with the MozBar

If you have a Moz Pro account (including free trial) you can easily find the most relevant keywords for any page on the web using the MozBar.

Simply navigate to the page you want to research. In this case, we'll use the MozBar to find related keywords for our page about SERP Features.

After navigating to the page, go to the keyword tab (KW) and type in your primary keyword that you want to rank for, in this case ‘serp feature’. The MozBar will then analyze the page and give you a list of the most relevant related keywords.

MozBar keywords

Adding these keywords and topics to the page is one of the easiest ways to earn higher rankings.

For a deeper dive into how related keywords help your SEO, Moz founder Rand Fishkin created a helpful Whiteboard Friday.

Next: Keyword Optimization

At this point, we've built out our keyword list and related phrases. But we will not rest here. It's time to turn these keywords into rankable content.