Keyword Analysis: Making Sense of Big Data

Techniques for transforming high-volume search data and keyword clusters into strategic SEO insights.

Speakers: Ozan Ketenci

Content author

Author
Ozan KetenciinSEO on Board

15 Jun 2017

Video Transcript

⏱ 9 min read• 1,530 words

Read the full speech or switch languages

Hey Zeo followers. Today I want to show you how to group keywords in your analysis. And how to make those huge keyword lists actually usable. First, let's look at how to build a big keyword list. The methods I use are right here on screen. I start with Search Console data. Whether we optimize or not, Search Console shows the terms we already rank for. Even better, we can check positions, impressions, and click-through rates. On top of that, organic traffic is all about competitors. You know this already. Look at what they've done. If we want the top spot. We need to know what the current number one is doing. So I use Semrush and Search Metrics. I look at the keywords competitors rank for. Then I check where we stand. Which ones are we missing? Which ones perform badly? Which ones perform well?

Finally, once those first two steps are done, I look for new keywords. What terms am I missing? What ideas haven't made my list yet? I pull those using tools like Keyword Planner, Ubersuggest, and Term Explorer. Of course, once you do all of this, you end up with a pretty massive list. Sometimes 10,000 keywords. Sometimes 15,000. Depending on your site size, you get an insane Excel file. Using that file and making sense of it can be really tough. To make this more usable, we need to group it first. And when I do that grouping, I follow the steps I've marked in orange here. First, I look at search volumes.

For smaller content sites, search volumes of 50-100 can matter a lot. But for bigger sites, optimizing for 50-100 is a stretch. Same goes for anything under 2,000, or under 1,000. It's a bit unnecessary. It just creates extra work. So start by filtering on search volume. Base it on your own site's size. You can also hunt for high-converting keywords. Are you targeting short tail? Or long tail? With short tail, competition is brutal, as you know. Search volumes are huge, sure. But so is the competition. You could be up against massive rivals there. Again, it depends on your site's size. If you want a more specific topic, go with long tail keywords.

Third, Google now puts a lot of weight on semantic algorithms. So I analyze keyword frequencies to do semantic grouping. There's no tool that works well in Turkey for this yet. So how do we analyze word frequencies? Say you have a list of 10,000 keywords. You drop that list into a frequency checker. It tells you which word shows up how many times. Just search "frequency checker" online. You'll find one easily. Then you can see the counts. And those top words become the main headings for your groups.

Say you run a keyword analysis around shoes. You drop it into the frequency checker. You might see "sneakers" come out. Or Adidas. Or Nike. Those become your subgroups. Same thing if you pull out Adidas. Run Adidas back through the frequency checker. Now you'll see the split between Adidas t-shirts and Adidas shoes. Fourth: search trends. In e-commerce, these matter a lot, as you know. Depending on what kind of site you run, you need a good nose for these trends. Sometimes searches like "how" or "where" signal learning intent. But words like "buy," "discount," or "installment" are different. If those show up in your keywords, conversion will be higher. That's what we call higher buyer intent.

Pick your target based on your site type. Then group your keywords accordingly. Finally, I use these five tools you see below. They measure keyword opportunity, difficulty, and competition. Keyword Planner already gives you a hint with its competition metric. Of course, that data isn't 100% accurate. Next to it you get suggested bids. Use both the bids and the competition score. That way you can at least gauge how hard it'll be to rank for that keyword. On top of that, tools like SEO Monitor and Semrush have dedicated keyword opportunity tabs. They show you the opportunity in a keyword directly. They can give you an actual number for it. Then there's Ahrefs and SEO QuickSearch.

Ahrefs is a web tool like Semrush, as you know. You can pull a keyword difficulty score right there. SEO Quick is an extension you install in your browser. With it, you run a Search and see the data inline. So for every result, position 1, 2, 3. All the way to 10, or whatever page you're on. It shows you that domain's authority. You can see how many domains link back to them. It even shows the old page rank. You get metrics like these, and way more easily. Sure, you could get all of it by running things through Majestic.

But with SEO Quick you get it all from a single Search. And seeing the data visually gives you a chance to interpret it. You go, look, the one ranking first has this many backlinks. I'm way behind that. Maybe I should focus more on backlinks, not just content. Or maybe I shouldn't target this keyword at all. It's just too hard for me right now. After all that, we had a huge keyword list. So we took them and grouped them. We've got tons of subgroups. And in those subgroups, we've got dozens of pages we want to target. So how do we target on those pages? You can't just dump keywords into the page and spam it. That doesn't work anymore, as you know.

So we have to develop this grouped data the right way. Then place it into the page. That's why I use these five methods. First off, especially on content sites, not e-commerce, related searches matter a lot. Related searches show up at the bottom of Google's results. Google gives you about eight suggestions there. It's basically saying, here are other searches tied to yours. It's based on recent searches on Google. Google updates that section itself. So what you see today might look different in 10 days. Or a month later. But you still use it when creating content. Say you type "how to dye hair" or "hair coloring." You'll probably see things like "how to get a hair color" or "how to dye hair.".

That gives you something useful. Say I'm building a page about hair dyeing. Then I should have a heading like "how to dye your hair." Sometimes the H1 is already obvious from those searches. Assume we're using our keyword in the H1. Then we can pick the other subheadings just by looking here. That's how important I think this spot is. There's also Google Autocomplete. It shapes our whole search experience directly. You type anything into Google. It completes it for you automatically. You can use search operators here. Say we type Adidas. Add a star next to it. It fills in the ending and gives us all those suggestions.

Or put the star at the front. If there's a word before Adidas, Autocomplete shows it to us. This gives us a solid idea of what Google sees as related. Synonyms and near-synonyms matter too. That's especially true in English. They hold a big place in Google's English algorithm. It's gotten much more important in Turkish too. We can find synonyms and near-synonyms for our keyword. Dictionaries are one way. We can also read forums about that word. Or we can go a few pages deeper in the results. Looking there gets us those near-synonyms too.

Google Trends is another tool we all use a lot. Google Trends shows us the yearly picture. What trend did your keyword get searched on over the past period? When was it really popular? When was it less popular? That's what this tool shows us. With Google Trends, you can target the right keyword at the right time. When I say monthly search volume here. You might think it's the same as those search volumes. But it's not. When you drop all your keywords into Keyword Planner, there's a second section. It's not for finding new keywords. It's for pulling search volumes, as you know. You can segment the data by month there.

You can see monthly searches for the last two years. January, February, March, all of them. What's this good for? Google Trends factors in everything, even five years back. So sometimes we can't see the real trend properly. But look at monthly search volumes. Say January to December 2015. Let's say one keyword peaked in September. It hit its highest point there. Then we can think about it this way. We should target that keyword in September of 2016 too. So we get to see that. I think seasonal shifts matter a lot in e-commerce. Right now it's back-to-school season. Check Google Keyword Planner yourself. You'll see how "school backpack" climbs. You'll see how hard it peaks in September.

Once you do everything I've covered, you've built a solid keyword analysis. Plus, with those final steps, you can actually put that keyword analysis to work. That's all I've got for now. See you in the next video. Take care.

This transcript was generated using AI speech recognition and verified against subtitle style rules.

Ozan Ketenci

Content author

Ozan Ketenci

author

Contact us

How we handle your data

Open as a full page