Using Related Topics and Semantically Connected Keywords in Your SEO

[Estimated read time: 8 minutes]

Back in February, we explored balancing keyword targeting with concept targeting. This time around we’re looking at using your knowledge of related topics and semantic connections in your on-page SEO processes. In today’s Whiteboard Friday, Rand talks about applying those ideas in ways that will boost your ranking potential and inform your keyword research.

Click on the whiteboard image above to open a high resolution version in a new tab!

Video Transcription

Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we’re chatting about using related topics and semantically connected keywords, not just for keyword research — although that is a potential use and application — but also for some of the on-page SEO processes that we run.

Now when I say “related topics and semantically connected keywords,” I’m not actually talking about the things that you would find through a traditional keyword research process. The idea here is not to say, “What other things are people searching for that I could target?”

This is really trying to define two different, unique kinds of use cases or extractions for keywords.

Those are: What are keywords (well, words and phrases overall), unique words and phrases that are used on more pages and paragraphs and phrases that contain the query you’re going after across the Web, and what are terms and phrases that are used by more pages that rank for that particular query?

This may sound a little technical, but it’s not too hard. Once I show you these examples, I think you’ll grasp it real quick.

Let’s say for example that I’m trying to target the word “food processor.” I’ve chosen that as a good keyword for me. It’s something that I want to rank for. I know that if I can rank for it, I’m going to do well. My keyword research is done.

At this point, I’m doing on-page SEO. I’m trying to make my page more relevant and my site more relevant. I’m trying to rank better for this, and it could be the case that using certain words and phrases on the page where I’m targeting “food processor” is very important.

Google might look at a page that’s ranking for “food processor” and say, “Gosh, it’s weird that this page doesn’t have this keyword, this keyword, this keyword on it. We would expect that a page that’s targeting ‘food processor’ should have these things.”

So I want to find two things. I want to find, in the top 10 or top 20 results that Google already has for “food processor,” what are words and phrases that are on those pages more frequently? And across the entire Web, the corpus of the whole Internet that Google crawls or at least the important parts of the Internet that Google crawls and indexes, what are words and phrases that are used more on those pages when the phrase “food processor” is present?

That’s what this chart is showing me. Essentially, these are things that are used more across the whole Web. These are things that are used more on pages that already rank for this term.

I’ve done this with two examples — food processor and rainforest. Rainforest keywords in orange, food processor keywords in purple.

For example, you might see that the word “recipe” is used across the Web on lots of pages that contain “food processor,” which makes sense. Lots of recipes that call for a food processor have the word “recipe” on the page. But those aren’t necessarily the ones that rank very well. So it’s over here. It’s high up on the “Yes, used across the Web” but low down on “Used by pages that rank well.”

Is it important to use it on the page that I’m trying to target? Well, maybe. It depends on how comprehensive I’m trying to be. Maybe I should think about targeting that on a different page, these kinds of things.

Something like — let’s go over to our example for rainforest — a word like “temperate rainforest,” which are less popular and commonly used both on the Web and in the results that rank than the more commonly thought of “tropical rainforests.” So Washington State, for example, near Seattle has some temperate rainforests, where you get lots of rain, but you don’t think of them as traditional rainforests. They don’t have like thousands of creatures in them. They’re not all hot and wet like they are in Brazil or Costa Rica or those kinds of places. So “temperate,” less commonly used across the Web and less common in the ones that rank well.

But something like “Amazon,” very common in things that rank well and in the middle of pages that use it and don’t. Many pages that use rainforests don’t describe specifically the Amazon rainforest, but many do.

Got it. Now what?

So now you’ve got this concept. What do I do with these? Well, there are really two big things that you can do that are pretty awesome.

So if I have a page that’s targeting rainforests, I want to think about: What are the topics and concepts, words and phrases that Google probably wants me to cover, that users and searchers probably also want me to cover? Those could be things like rainfall, ecosystem and biome, tropical, Amazon like we talked about. Maybe even a competing brand, like National Geographic, which is on here. It’s used on a lot of pages that rank well. Maybe Google has an association between rainforests and Nat Geo, and I should potentially reference them or link to them or talk about them, pull a photo from them, that kind of thing. Brazil.

These words, using them on the page can help me to be more relevant, more comprehensive, potentially more useful, and more high-quality. This is especially true for informational style searches, but potentially true for commercial searches too.

Instead of just saying like, “Hey, I’m going to look for things that people also search for. I’m going to use Google suggest and related searches. I’m going to use KeywordTool.io, or I’m going to go Google AdWords and see what are the other high-volume searches.”

I might broaden my thinking to, “Huh, I wonder if things like ‘food processor recipes,’ or very specific things, like ‘pesto made with food processor,’ are interesting things for me to target additionally deeper in my site so I can build authority around all the topics and concepts that are related to the word ‘food processor.'”

Not every one of these semantic and related topics is going to be a good choice for you. That’s definitely the case. You have to use good judgment and the traditional metrics that you would use for keyword research — volume, difficulty, opportunity — to discover the right ones.

What’s kind of cool and one of the reasons I’m covering this, this week is that some tools have come out in the recent past, a bunch of NLP, Natural Language Processing tools, and APIs that let you do some cool stuff around this. Those include people like Alchemy, Sysomos, OpenCalais, and a number of others.

Then it’s also the case, and this is slightly self-promotional, but Moz Analytics [Moz Pro] recently released their Related Topics feature. So you can now go to the on-page section of Moz Pro and see a list of things. The Moz Pro one is going to be more like the stuff here. Think words and phrases that are used by pages that also already rank for the query you’re targeting. Then, in about a month, Keyword Explorer will be launching, which I’ve talked about a number of times, and that will have more of these things. It’s used on pages across the Web that also feature this.

But you can get this stuff currently through some of these tools. You can do your own analyses. There’s lots of code out there in code repositories that you can pull from the Web. So I encourage you to give this a try. We’ve seen some good results from people who are trying this stuff out, who are including these terms and phrases, and who are broadening their keyword research with it.

Look forward to your comments, and we’ll see you again next week for another edition of Whiteboard Friday. Take care.

Video transcription by Speechpad.com

P.S. Another tool that offer this type of functionality is the excellent SEOZoom suite — currently Italian-language centric, but very powerful (I got a demo at EMetrics Milan that really impressed me).