English Español Deutsch Français Italiano Português (Brasil) Русский 中文 日本語

SEMrush Toolbox #5: Keyword Gap


Click here to get Jake's magic spreadsheet.



Craig: Hi guys, and welcome to today's SEMrush webinar. It's the fifth in the series of the SEMrush toolbox. Today, we are going to be talking about the Keyword Gap tool. Basically, what the Keyword Gap tool allows you to do is check up to five competitors, and you can use your own website, as well, and just find out opportunities of keywords your competitors are ranking for compared to your own site.

Today's guest is Jake Bohall. Welcome online today, Jake, and for anyone who has never come across you before, do you want to tell the audience a little bit about yourself?

Jake Bohall: I'm vice president of marketing at Hive Digital, which is a digital marketing agency focused on companies that are working towards a better world. I've been doing SEO for over 10 years, and I really appreciate the opportunity to speak with all of you. Thank you to SEMrush for having me.

Keyword Gap Overview

Craig: Perfect. Thank you for that. What we'll do now is we will quickly just jump into the webinar. First off, I'm just going to give you guys a brief overview, so I'm going to start sharing my screen.

What we're going to do is just show you the Keyword Gap tool and what it does from a basic point of view and then Jake's going to come on and give us some insights and stuff like that as to how Jake uses the tool and hopefully some good knowledge and information that you can implement into your own strategies.

The first thing that's really important, and I know it sounds really simple, but a lot of people in the past that are watching these tutorials tend not to know where the tool is, so on your dashboard on the left-hand side here, we have Gap Analysis. If we open up Gap Analysis, you've got Keyword Gap, Backlink Gap, Bulk Analysis.

We're going to be talking about Keyword Gap tool today. I'm going to click on that. What it does is bring you up a nice big dashboard.

What is the keyword gap tool? As I said at the start, you can compare domains. I’ll put my own domain name in here, where you can just enter your domain name, and I'll put in a couple of competitors, so matthewwoodward.co.uk, and I'll do Andy Drinkwater. I've put in three different websites there.

What's quite important for you to notice is you can look at the UK database, so you've got an option right here where you can change that to whatever database you're operating in, so make sure you do look at that. You can look at from desktop or mobile, and obviously, you've got currency and stuff that you can change for the cost per click stuff. That's not hugely important.

What is really important is that under my domain name, I can compare my organic keywords. You've got three options thereunder organic keywords, and that's organic keywords, paid keywords, and PLA keywords, which is Google Shopping.

I can compare my organic keywords to Matthew Woodward's paid keywords if I wanted to compare what's going on. The Keyword Gap tool doesn't just work for organic search. It's something you can look at. If someone's doing really good paid and they're spending a lot of money, and getting some good keywords that convert well, you may want to implement them into your organic strategy or vice versa. You can compare organic with organic, organic with paid, or organic with Google Shopping.

Over in the right-hand corner, you've got a big green Go button. You simply click the Go button and it will pull all the data for you.

So, you've got all keywords, unique to the first domain keywords, common keywords, and, unique keywords that are there. You have to make sure that you're ticking the right box, or you're not going to get any data. You can change that across the different domain names. Make sure it's all keywords that you do select.

When you do that, it brings up the nice table, and it will tell us all the keywords there that I rank for, so “SEO consultant”, I'm position 35, Matthew Woodward doesn't rank for it, Andy Drinkwater's position 66. Obviously, you can work through all of this data. It'll give you the volume and all the other data there that is going to help you establish whether that keyword is important.

For me personally, I'm always looking at volume. I don't feel that there's any point chasing a keyword that gives low or no search volume.

Obviously, other people look at various other things, such as how competitive that term is and stuff like that, or whatever suits your own niche, if you like. You do have advanced filters right here if you further want to drill down the keywords and stuff like that, as well.

You can also include, exclude, specific keywords, websites, volume, or use the containing word match, exact match, and begins with or ends with. You can add more filters in here, just keep adding lines, click apply, and that will allow you to then use advanced filters on the data that's provided. That is quite good.

You can also enable a chart view if, for any reason, you want to share a chart with a client or whatever it may be. Sometimes, people feel that charts look nicer than a table. You can, as normal, with most of the SEMrush tools, export the data, as well. You can export all this information here if you want to relay it back to a customer, get their take on it, or whatever you're doing for the client. You may only be doing consultancy and this may be part of your process.

As I say, this tool is very simple and easy to use. The organic keywords against someone's paid search, I think that's quite important. A lot of people think it's just all about organic search. Even if you want to check your paid search compared to someone else's paid search, this is the perfect tool to do that.

You can see there, there are lots of opportunities for me to grab stuff that Matthew Woodward ranks for and implement it into my own website, for example, “how to start a blog”, if that's something that I felt was going to make money for me, then that opportunity's there. You can see there's a whole ton of keywords going down there. You can see the little dash there, which means I don't actually rank for it. These are all opportunities for me. A lot of these are getting thousands of searches per month.

Everything's there for you, it's laid out on a plate, and just finding these opportunities very quickly is ideal. There are people out there who do manual research, or they'll just check people's blogs out and just guess, or give a halfhearted attempt at trying to replicate someone else's strategy. As I say, with SEMrush, they're bringing out new tools all the time, and these things just make life a hell of a lot easier.

It's a very simple and easy-to-use tool. It's not something that's overcomplicated. It doesn't do much else other than this. Really importantly, make sure that when you're looking at it, this little icon here, you do click to make sure that you can see all keywords. Obviously, if you then want to filter that down, you can mess about with that filter. Have a look at that tool if you guys are doing keyword research and you want to check out your competition and see if there are opportunities for you.

We are now going to get Jake to come on and go into things in a bit more detail.

Keyword Research: Developing An Actionable Keyword Strategy

Jake Bohall: Awesome. As Craig was pointing out, the tool itself is very simple to get started, but a lot of times, what happens is you get to that point where you have 3,500 different keywords, or 10,000, 20,000 different keywords. Being able to distill that down into an actionable strategy takes a little bit more effort. That's part of what I'm hoping to walk you through today.

As Craig mentioned, the Keyword Gap tool. Put in the different domains for any competitors that you're trying to compete against. When we're looking at this, we want to think, "Why are we doing keyword gap analysis? What benefit does that have for us?" I use it for prioritizing opportunities for new content, keyword discovery, and evaluating performance versus a competitor.

At the end of the day, the Keyword Gap tool is meant to help you identify those areas where everybody else is looking great, they're covered, and you're not. Where are we missing out on opportunities? I'm going to walk you through a process that I use to take an export of all that SEMrush data and bucket it into these seven different categories to help me prioritize which keywords should I be going after, which ones should I not.

You'll notice, in this case, there are about 10,000 keywords that I'm evaluating here. Obviously, we could do more. If you have some sort of limitation in SEMrush on the number of rows that you can export, a little bit of a pro tip would be to go ahead and slice the segments for keyword ranks for the first only, then do an export, then ranks for the second only, then do an export, and you can combine all of that data into a single spreadsheet later.

If you were to go and do this particular search here for these five domains, I picked Score.org, the Small Business Administration, Kauffman, MicroMentor and so forth, these are all websites that focus on entrepreneurship, helping people develop and grow their business and that type of thing.

You'll see that at the end of the day, there are 454,000 different keywords. We're only going to look at 10,000 of them today, just so that my spreadsheet doesn't explode on us. There are ways you can automate this process, and I'm doing it on a smaller sample size, but the same rules apply, and it would go for all of the data.

I'm going to switch over here to my spreadsheet... When you do an export, you end up with this sort of list we're going to get to here where it's filtered, color-coded, and all that sort of fun stuff.

The first thing we're going to do is we're going to export the content to Excel, and then we're going to work on actually filtering that data so that we can manipulate it. There's a couple of things we have to do to fix some formatting, and we're going to do some highlighting in order to just get things prepped for use.


Excel Hacks For Formatting Keyword Data

This is the raw data that comes out, the raw export that comes out. You'll see, in my version of Excel, I end up with these weird issues where Excel thinks it's a number but it's actually stored as text. The first thing I do is just go to Data and then Filter. This is so that all of the rows stay intact whenever we're making changes or filtering by certain things. Then I'm going to do a little trick where we do Text to Columns and hit Finish, and it just converts that into a number format.

That makes it easier for us to do sorting, filtering, and that sort of stuff later. We just quickly go through and convert all of these into a number format so that we can manipulate it. It's literally just hitting Text to Columns and then finishing, and it's going to convert it. That's the first.

The next thing we're going to do is we're going to go ahead and set up some conditional highlighting. This is going to help us for sorting later. We're going to take this entire section, our site, in this case, we're assuming that we are Score.org, and all of our competitors, and then we're going to just do some conditional formatting.

We're going to create a rule where if the ranking is between positions 1 and 25 we want it to be green, and then we're going to add another one where if it's position 26 to 50, we're going to highlight it in red. Then, immediately, out of the gate, we can quickly see everything where our site is in the top 25. We'll be able to do those kinds of filters quite quickly.

The last thing we're going to do is we're going to add a new column here, we can call it a priority, we can call it a bucket, whatever you want. This is basically going to be some groupings that we're going to establish. Those groupings that I showed you a moment ago, like attack, block, content creation, or whatever they are, those buckets are going to go here.

After this, we're going to go through a process of bucketing the terms in which only one site is ranking. We're going to prep the data, and then we're going to bucket terms with only one site ranking. This process involves taking and filtering all of these results. We're going to filter them to zero. We want to see where everything is a zero, with the exception of one site.

This is going to help us find some potential edge cases. If I just pulled all of this data and I can see that Bplans.com is the only one that has exposure for this (keyword), that either means that maybe it's a brand term, or maybe it's some kind of a weird outlier because that keyword isn't transcending all of our competiтп websites.

We do this process for each of these competitors.

At this point, what we're going to do, just so we can start to get rid of some of the cruft, we can go ahead and set a rule here that we only want to want to look at blanks. This is essentially going to allow us to continue to shrink our list down as we move along. We'll do this for every competitor.

Finally, we get to the very end, where just our site is the only one that's left. I don't want to take up time just doing this over and over again, because it is somewhat repetitive, and, again, a process that should probably just get automated.

Ultimately, you're going to get to that same section for yourself, we'll filter when our site is the only one that's ranking, and we'll do that same process, except for in this case, when it's our site that's ranking, and we see that it's over 50, odds are, this is going to be an outlier for us as well, so we go ahead and just mark those as unlikely.

That's the first step of this process, just going through and making sure that we have a filter established for each one of these. That allows us to get rid of areas where only a single site is ranking. I mentioned you go through for one competitor that's the only one that's showing up, then you run the filters, and then you start back over. You do that for all of the competitors until you've done all of them, and then you make this change when doing it for yourself.

Segmenting Keywords Into Buckets


The next thing we're going to do is we're going to start bucketing terms where we feel like the competitor has significantly improved visibility compared to us.

For here, what we're going to do is we're going to filter ourselves to red so we know, "Okay, these are the ones where we are ranking in positions 26 through 50," and then for each competitor, we're going to go through and filter them to green. There may be some, where like this case, Bplans is doing really well, and we're doing okay, this is an opportunity where we should be going after them.

This we would go ahead and set as an attack. There's going to be a smaller bucket for these.

Then we're going to go and look, now, at terms that we should be protecting. These are phrases where, generally, you're ranking in the top 25 and the competitor is also ranking in the top 25. Again, we're looking at bucketing, in this case, any areas where our site is green, and then we'll pick if this site is green, so we're going to protect. We just go through this process for each one of these.


The next group that we're going to look at is what I refer to as catching up. These are going to be when the client is ranking greater than 25, but then our competitor is ranking in the top 25. Our goal here is to create some categorization so that instead of trying to process through 10,000 keywords, we can quickly get to a list of what is actually important.

Next, we're going to look at bucketing keywords where it appears that we're under the radar. These are the phrases that we alone rank for and the competitors are not doing well at all, so we're under their radar. Obviously, that's something we need to be mindful of because they could come after us. Sometimes, this is going to be our own brand terms, custom content, or things that we have. Other times, this is going to be actual keywords where you're owning it and the competitor is not.


The last couple of buckets we're going to hit, the first one is going to be outliers. We're going to look at where the client and the competitors are all not performing well. Some of these could potentially be interesting, but we're not necessarily interested in prioritizing them compared with everything else. Again, part of the goal here is to create a quick bucketing system for all of these keywords so we can then come back and review them later.

The next item we're going to look at is what we call unlikely. These are phrases that definitely need to be manually reviewed, but they're often brand variants or completely unrelated terms that need to get disregarded.


What will happen when you're done is you'll end up where all of your keyword phrases are going to be assigned into these particular buckets: Attack, protect, catch up, potential, under the radar, outlier, and unlikely.

A lot of times, what I will do is I will take the attack group, and now I've categorized these as keywords that we definitely need to be going after, and then I can prioritize them using the rest of the data that was originally exported from SEMrush. I can sort by search volume and see that these are an obviously higher priority, we can look at them in reference to cost per click.

That same concept applies to the things that we should protect. We're doing very well for a keyword like “business plan software”, but we have some people coming up behind us. In this case, we have someone that's beating us. We need to really hunker down and put up a much more focused effort on preserving these because this is where our money's coming from right now.

When we're looking at the catch-up bucket, these are areas where the competitors are seemingly doing better than us and we're completely missing the boat, so things like income statements and how to start a business, we're missing all of that.

Inside each of these buckets, you can start to slice by search volume. Craig had mentioned that you don't necessarily always care about keywords when there's five people a month that search for it. If you're trying to do priority, once you have these buckets established, you can then go in and start to prioritize within them.

Keyword Bucket Terminology Explained

Just for the sake of some explanation, the attack group, those are the keywords that I felt represent a direct opportunity for you to edge out your competitors. They're performing well, you're on page three to five, so it's something you should enter that market in attack mode.

Protect are the ones where you're in the first couple of pages, but your competitors are also there, or they're collectively gaining some ground on you, and you want to make sure that you're hedging yourself for any future growth from them.

Bucket three was catch-up, which are phrases the competitors are ranking and you are not. These are things where we feel like most of them are going to be low-hanging fruit terms. A lot of times, it's just a matter of including those keywords in content or building new content around them, and immediately we start to get some exposure. That's because generally, they're thematically relevant to our site already and they were just opportunities that we missed.

Four is the phrases where one of our competitors ranks really well, or several of them, and we do not. We put these in a separate bucket because it's something where they're likely relevant but there could be some mismatch where... if multiple competitors that we're pulling are offering a particular business service that we don't offer.

Number five is under the radar. These were the ones where we were ranking, the competitors weren't. Keep an eye on them. If you see competitors coming in, know that there's action that you need to take to keep ahead.

The last ones, outliers and unlikelies, are ones that you should probably take a look at. The unlikelies are the last bucket to look in, and the outliers are the ones where the competitors seem to be ranking well, but there's not really a clear winner. Either the competitors have a high-ranking position and nobody else does, so it just looks weird, or it's an area where several of the competitors are ranking in position 50 and 60. Maybe there's some intent because it seems there must be if multiple companies are ranking for the same phrase and we aren't. It's something that warrants getting a closer look.

By going through this process, obviously, we can bucket these types of keyword phrases into these groupings. We took these 10,000 keywords and said, "There's 68 that we can prioritize as attack terms, and these are ones where we'll probably want to go out and do some more development or other strategies to really pursue them more aggressively."

The protect and catch-up terms, those are the ones that we immediately need to send off to our content team and say, "I want you to look at the page that's ranking for this particular keyword phrase and figure out how we can bolster it." Potential are terms that we might want to go after, and there's a lot of search volume available, but we probably need to do a little bit more manual review.

We just took 10,000 keywords and, in just a matter of minutes, have identified several buckets with 430 keywords that our content should be working on, 70 keywords that our link-building team should be working on, and then now, for the keyword discovery part, I have a much smaller list, a more manageable list that's grounded in evidence of our competitors getting exposure for those keywords in order to come through for more keyword discovery and that type of thing.

Ultimately, if you line things up, you're going to win. That's the goal of what we're doing here.

Craig: I think if anyone wants to grab that spreadsheet, if you reach out to Jake on the email he mentioned earlier, on Twitter, or whatever, you're best going to the man direct himself, and obviously, SEMrush and whatnot will share your slides and stuff like that from the presentation, as well.

Guys, we've got six minutes for questions and answers. First one: "Once you get all the data sorted, how do we develop content or adjust current content to capture those keywords?"

Winning Your Competitors’ Keywords

Jake Bohall: Once we have the bucket established, oftentimes, I’ll see which pages are ranking. Is it my target page that is ranking for it? Do I have multiple pages that are ranking for it? It's a matter of identifying, am I including this as often as I could, or as well as I could, I should say? It's not a matter of the count. A lot of times, when we were pulling this, the things that we've done well and the content that we've built out well are generally the ones that are in protect mode. The ones that are in the attack buckets are ones where we haven't really built out the content that well, they're not established pillars, there's not a lot of resources around on our site, we don't have a lot of links coming in.

It's this starting point of saying, "What have I done to fulfill the user intent? When I do a search on Google, what is showing up there that I'm missing? Are people looking for charts? Are they looking to buy? Are they looking for more information? Are they wanting to get a comparison or what have you?" Once you have the buckets, it is a more manual process of actually going through and figuring out, "What is the gap in the actual content and intent of the page that I would like to have ranking and those that are out there now?"

How To Protect SEO Rankings

Craig: Great answer, perfect. The next question: "When you mark your keywords as protect, so you want to protect those rankings, what are you doing to protect and improve those rankings? Is it just content, or is it links and everything else that you believe works?"

Jake Bohall: Both. They're going to need ongoing maintenance. I don't want to ignore them just because we seem to be performing well. A lot of times, it's a combination of trying to share or present the content better to the community so we can try to get some links back to it. A lot of times, we'll see we're in position five and six, a competitor may be in four, another one in nine, and that's why it ended up in that bucket. It can be content, it can be links. There's a number of different techniques that we can take to it, depending on where it's efficient.

My goal with these buckets is to categorize so that I know and can hand-off to the appropriate teams internally. The ones that are needing new content, I know I can immediately send that off to the team and say, "Go find out why we don't have very good content for this because we're obviously not performing well." Those that are in the protect or attack mode, I know that needs to go to a content team and a link-building team, and odds are, we need to be hands-on in identifying a specific strategy around each of those.

There is no automated system that perfectly says, "This bucket, this is exactly what you do." It's about getting them into buckets so that I can have my mindset on, "For these terms, I know I need to do some authority building, I need to do some relevance building, and I can sit down with the team that I know does both of that, and we can be more effective at coming up with strategies, and we're not spending times on outliers and that type of thing."

Keyword Density

Craig: I think that's a sensible way of approaching it. Can you talk about keyword phrase density?

Jake Bohall: I wouldn't put forward any idea to say that a keyword density of any specific number is more or less appropriate. If it's something you're interested in doing and you think that it may be important for your keyword space, my recommendation would be to evaluate the sites that are ranking in the top 10 and look at their density around that particular keyword. If you want to use that as a benchmark to say that's what's normal and that's what Google is accepting, you may find a range.

Craig: You have to make the content focused around a particular keyword without doing repetitive spamming or whatever, but as you say, not enough just to mention it passing by. It's just getting those ... the balance right or whatever. I think looking at your competitors is important.

Sadly, due to your excellent presentation, we have no more time for questions. People, if you do want to ask any further questions or get that spreadsheet, do get in touch with Jake. I'm sure he's got a lot more to offer, more tips, and stuff like that. Sadly, we're out of time today. Thank you very much for being on today, Jake. It's been a pleasure.

Jake Bohall: Absolutely.

Craig: We will be back on in a few weeks with another SEMrush toolbox webinar. See you all then. Cheers.

Jake Bohall: Great, thank you.

Craig: Bye-bye. Cheers.

Jake Bohall: Take care.

All levels
Discover SEMrush