When it comes to keyword research, I personally love Google Search Console (GSC). But, as an SEO expert, you probably know that you cannot fully rely on stats from the Google tool. Why?
Because Google Search Console comes with a lot of limitations.
First, it only features impressions — as in how often your link was seen. This means that you can depend on GSC intel when your page is already ranking within the first 10 Google search results. If not, you’d have to turn to the concept of search volume.
Secondly, GSC shows keywords you are already ranking for—which means that it can be perfect for analyzing your own site’s stance in terms of keywords. But if you want to expand your keyword portfolio and reveal any keyword gaps between you and your competition, it’s third-party tools that come to save the day.
With this in mind, I’ve decided to bring to your attention my absolute favorite external tool for keyword research—Semrush.
For the past few years, they’ve been expanding their keyword database, which is now considered the largest on the market.
But while, in this case, database size matters, we all know that keyword stats go way beyond pure scale as keyword research really boils down to the quality of data you have on hand to analyze.
Now, let’s see if Semrush managed to turn quantity into quality and what research has to say on this.
Why Not Google Keyword Planner?
Before I move on to discussing the ins and outs of Semrush’s keyword database, I can anticipate one question — why not take keyword search volume data from another Google-native tool, Keyword Planner (GKP)?
Well, first and foremost, it’s a tool for managing your paid campaigns, so it’s not always best for SEO purposes. For instance, as it has to follow certain Google advertorial policies, medicine-related keyword data is just missing.
Plus, GKP can be simply confusing as it doesn’t always offer the cleanest data. For instance, its search volume stats combine numbers for search queries that have a similar meaning or plurals, so you can’t always check specific stats for specific queries.
Why Opt For Third-Party Tools?
This all leads us to address why is it that I’ll be talking about Semrush specifically because it’s not the only platform offering intel on all things keywords.
The thing is, third-party tools all take their insights from clickstream data topped with proprietary AI algos. This means that the more data suppliers they have and the smarter machine learning algos, the more accurate insights you receive.
That’s why analyzing and benchmarking each tool individually is the only way to actually assess the data accuracy and quality of all the estimations.
As I mentioned, Semrush has the largest keyword database compared to all the other solutions. But while numbers don’t lie, quality assessment is a different story.
Semrush recently announced that they made enormous improvements to their algo for search volume predictions. How so?
- They grew the number of data sources by 5X (and if you know how machine learning works, you’ll know that the more data you have, the smarter your algo).
- They added a few more algo types to their overall AI ecosystem, enriching it with Natural Language Processing (NLP)—it’s what Google itself uses within its search algo— and Anomaly Detection algo that helps to exclude all the exceptional data to avoid inaccurate assessments.
Plus, and I know this from my own experience, Semrush is doing a great job at getting rid of obsolete data within their keyword database.
Many other solutions don’t get remove old and unnecessary data (think FIFA 2016-related keywords) from their databases just to showcase scale, but that actually doesn’t add up to quality.
Third-Party Tool Comparison for Keyword Data Quality
Well, now that you have all the background, let’s get to hard numbers. While Semrush may be claiming a lot of cool features about its database, we can’t just rely on their word.
That’s why I’d like to show you the results of one study that stacked Semrush’s keyword data stats up against that of other solutions—Google Keyword Planner, Moz, Mangools, Ahrefs, Serpstat, and Sistrix.
The key benchmark was impressions data from Google Search Console as in some cases—and this is what the study took as a reference—impressions can depict the real search volume data. You can read up on the full methodology here.
Search Volume Quality Compared
Initially, we see a more simple pure number comparison when all the keyword search volume data from various solutions was stacked up against that of GKP.
How to read the graph below: The higher the percentage, the more match between Google Search Console’s impressions and the tool.
Semrush has an obvious win here as the second-closest “competitor” is Google Keyword Planner, but you can disregard it as—as mentioned— it only features value ranges, that’s why its numbers would look inaccurate.
Now since Semrush came out as a clear winner, we also see a tool-by-tool comparison with Semrush—just to see the extent of the accuracy difference.
Almost in all cases, Semrush returned at least 30% more accurate search volume data, which in terms of building an SEO strategy can be that fine line between success and failure.
Apparently, the biggest cause of this data difference has to do with the way solutions reflect search volume data for the so-called trendy keywords‚ or search queries that are just gaining their momentum across search and we don’t yet have years-long data to analyze them.
So if you are in a business where you might have to use and, most importantly, show up for the most recent trendy searches, you cannot afford to rely on inaccurate keyword data.
Keyword Data Coverage Compared
Another aspect of keyword volumes that point to data quality is coverage. While a tool can report that it has a billion-, or even trillion-keyword large database, you need to know how many keywords it actually does not cover.
That’s where you analyze the number of keywords for which there’s no data (typically keywords that come with a “not found” label).
The study analyzed this, too.
How to read the graph below: The smaller the number, the fewer “not found” keywords you see.
As it appears, when it comes to keyword data coverage, Semrush wins once again.
And in this case, the margins are even larger. The second-closest competitor in this regard (if we exclude Google Keyword Planner’s free version) was missing data for twice as many keywords as Semrush.
Why Keyword Search Volume Is the Most Important Metric For Your Keyword Research
Now that you have a good understanding of the keyword tool landscape and know how each tool compares to another, you might be wondering why focus on search volumes?
Every post I have on keyword research outlines the general logic of the process:
1. Create a list of relevant topics that you want to rank for and identify your seed keywords
I always suggest narrowing things down to 5-10 generic topics. The pages that cover these topics are the ones that you set up for specific keywords.
At this stage—very early on—search volumes already come in.
Every topic you’ve picked implies a list of seed keywords.
Say, you’ve pinpointed that you want to rank for “cleaning services”, well, your seed keywords would often repeat these words. Your seed keyword list could look like this: “cleaning services”, “home cleaning”, “pool cleaning”, and so on.
But the catch is that you can’t just pick any keyword as you can’t possibly create hundreds of pages all at once. You have to go smart about your keyword choices so you’ll have to filter your list by a few metrics, with search volumes being the key factor.
On top of search volumes, you’ll also want to look at:
- Keyword difficulty—a metric that reflects how many rivals you might be competing against if you want to rank for a given keyword,
- Competitive density (shows the level of paid search competition), and
- Keyword intent that unwraps what stage of search you are tackling: informational keywords are perfect for targeting users at the awareness stage, and transactional keywords tackle users who want to complete a target action (purchase your product, order your services, etc).
All these metrics can help you not just to focus on keywords that can potentially bring in the traffic, they enable you to use keywords that you can realistically rank for. Because if you try to target highly competitive terms, you’ll likely see little to no results. At least, initially.
2. Explore related keywords for additional value
Once you have your list of seed keywords, you can start expanding it with similar target terms you better also use across your content.
Because keywords are what you’d call an ecosystem—if your page is targeting “pool services”, you can’t simply place that keyword across the entire page and expect high rankings.
Google knows that a high-quality page that talks about pool cleaning services, should also include words like “swimming,” “acid-wash,” “pebblecrete pool,” etc.
And once again, you cannot squeeze in all possible related keywords into one page. Search volume will often define your choice because you should try to target keywords that have the widest reach in terms of monthly searches.
3. Go long-tail
With that expanded keyword list, you should try to turn some of the generic keywords you have picked into long-tails as they typically imply a more commercial nature.
If someone looking for “pool cleaning services” might simply want to find out how to start a pool cleaning company, a person searching for “pool cleaning company in San Jose” is more likely to want to hire a company for cleaning their pool.
Plus, long-tails are typically less competitive than shorter search queries so you might see faster results.
If you manage to come up with a large list of long-tails, you can once again turn to search volumes to pick the most impactful search terms. But at this stage, you might not have a wide range of keywords so don’t completely disregard terms that imply lower user interest.
4. Finesse your final keyword list
At this point, you should have a pretty large and impressive list of keywords you should integrate across your site.
But I always suggest taking another look at your entire list—if you have more than 500 search terms in there, save your time and effort by prioritizing your keyword implementation. How? Search volumes will help you once again.
Simply sort your fina master list by search volume and take the first 50 keywords with higher counts for this particular metric. Because at the end of the day, your pages should generate traffic.
What’s next? You create or rewrite your existing content with these keywords in mind.
I have an entire guide that reveals how to write content people will love and share. Because while smart keyword research is the very foundation of your content strategy, you should still focus on your reader and invest in creating engaging content that resonates with your audience.
Is Semrush Right for You?
As you know, I don’t like to take anyone’s word for granted. That’s why we’ve gone through all the ins and outs of keyword data quality and reviewed search volume stats of the most major solutions on the market.
As you can see, Semrush had an uncontested win in each category, making it the most reliable solution for keyword research.
And you know what makes me vouch for Semrush on top of quality data? They are actually constantly working on improving it despite moving further and further from the competition each year.
Just recently, they announced that their keyword coverage for the US grew by 19%, which brings it up to 98% coverage of actively searched queries in the US.
That’s almost complete coverage we’re talking about!