Ahrefs Keyword Explorer Tool – Why My Mind Was Blown By Tim Soulo

Most people in the internet marketing space have heard of Ahrefs. Among those, a pretty large percentage of them uses it for backlink checking and some of the other key features.

Not that many people use it for their core keyword research though. I've used the keyword tool and found it to be pretty good, but it wasn't until I attended a recent presentation by none other than Tim "Mr Ahrefs" Soulo that I realized just how exceptional the Ahrefs keyword tool is.

Prior to attending the speech, I didn't see much difference between the main keyword tools in terms of the bigger picture. Long Tail Pro has some advantages, SECockpit has some, Ahrefs has some. Whichever you use doesn't really matter, because they're all using the same data and they all have similar features.

Or so I thought...

It turns out, Ahrefs has a far superior tool, that I'm going to be focusing the majority of my keyword research around in the future.

Tim's kindly given me permission to share some of his slides from the presentation with you, and so over the course of this post, I'm going to teach you the following:

  • Where Ahrefs gets its data compared to other tools, and why that matters so much

  • Why the Ahrefs "traffic" score helps you find the keywords everyone else misses
  • How we're going to use Ahrefs for our niche sites moving forward.

Where Most Keyword Tools Get Their Data

Until recently, I thought all keyword tools got their data through the Google API, or by scraping the Google Keyword Planner (GKP.) I wasn't sure exactly how it worked, and it turns out, there isn't a Google API, all the tools just scrape the keyword planner, which seems a bit off.

The problem with this is that Google doesn't really ever give the whole data. They're not supposed to, because the whole purpose of the GKP is to sell advertising, and not worry about the long tail keywords without many searches.

To illustrate this issue, Tim pointed out how for the word "iphone", Google only shows 590 possible related keywords.

Seriously, only 590 variants or keywords related to iphone??!

Now, most professional tools do a better job of this though. Semrush for example, shows 850,000 related keywords, which is much better.

But Ahrefs shows over 3 million!

We like people who like memes.

It's simple, using Google as the source of your keywords isn't going to get you the full data. Sure, you could scrape the whole Adwords database and you'd get a ton of data...but the Adwords database doesn't have everything, it's aimed at advertising, not SEO after all.

Also, Google Keyword Planner finds all related keywords and adds them up to get the total search volume, rather than showing the individual volume for each variant. In many situations this is a bad idea, since you'll think some keywords get far higher searches than they really do.

Look, even among Google products there is no real correlation:

This side shows that even three of Google’s own sources differ on search volumes. Accurate much?

So where does Ahrefs get its data from then & why should we care?

Ahrefs Uses Clickstream data - Here is why it matters!

When Ahrefs relaunched a lot of their services a little while back, I didn't pay much attention to how good clickstream data was. I remember them talking about it, and I thought "Yeah cool, sounds nice but how much does your tool cost now?" which was a bit shortsighted.

Remember though, I thought all these tools were pretty much the same.

Ahrefs buy as much clickstream data as they can, and then they model that data against what the Google Keyword Planner shows them. With GKP as their base, Ahrefs are able to utilize clickstream to make better estimates of search volume.

Aside from Moz, no other tool uses Clickstream data to calculate keyword lists and search volumes, which is why Ahrefs ends up with a much more substantial database.

Why does this matter? Because while Ahrefs still uses a 30-month average to calculate search volume, it still bases these averages on real data, rather than whatever the Google API wants you to see.

One thing Tim wanted me to point out is this: Clickstream data probably only accounts for 10-20% of internet users, so it's impossible to determine whether ahrefs is more or less accurate than the GKP. They're definitely more substantial, and have better estimates though.

As an aside, if you ever wondered why "Keyword 2017" shows lower search volume than "Keyword 2016" even though we're near the end of 2017, that's the 30-month average at work. In reality the latest year would almost always get more searches than other years, which would eventually drop off the database.

Quick update: Since I wrote this, Ahrefs has rolled out a new "Newly discovered" keyword feature, which shows keywords they've found in the past few months, depending on which plan you have. This is great for finding new/trending keywords and getting accurate search data.

That's not the only mind-blowing aspect of the using clickstream though. It gets better.

Ahrefs shows the most accurate "traffic" volume of any keyword tool I've seen, and that's because it uses a model combining Google Keyword Planner + clickstream.

Every tool out there tries to predict how much traffic you'd get for any particular keyword, but they're basing this off estimates (which are also based off estimates as well.) The reason Ahrefs is better is because their model for creating estimates uses more than 1 data source.

For example, it's relatively well known that in any SERP, the website in position 1 would get the lion's share of the clicks. Tools like Semrush will guesstimate how much traffic you're getting by looking at your search position, the search volume for your keyword, and then giving you a % of that. Ahrefs also does estimates like this, but thanks to clickstream, they can see the real CTR curve of every keyword, rather than applying general industry averages.

Considering that every SERP is different, every meta description different, and every user different, it's a helpful, but flawed model to just base traffic stats on estimates.

Ahrefs bases their traffic stats on their model, which is based on two things:

1.) Actual Clickstream data showing who clicks where,

2.) Combined traffic data for all related keywords as well.

Remember, it's still an estimate, just a better one.

This slide shows that despite a higher search volume, the first keyword has far fewer related keywords and far inferior total traffic.

How Does This Help You?

In more ways than you may imagine.

Look at the slide above, and you can see that a keyword with lower search volume actually records higher traffic. 

In this particular case, it is all about the popularity of the topic in general.. and the uniqueness of the search queries in that topic. Squeeze page is a super unique term.. there's no way to search for it in a different way.. while "how to drive traffic" is a very generic search query, there are tons of ways to search for it..

You can use this to your advantage. While other niche marketers are going for the keywords with the most hypothetical traffic, you can slip underneath and optimize your content for the keywords that get the real traffic.

If you combined this with something like Doug Cunnington's KGR method (which I've done), you can see how quickly you can build up long tail traffic without having to go out and compete with everyone else.

I'm not saying you should completely ignore the other keywords, but when it comes to seeing which keywords to build your campaigns around, and which ones to prioritize, you'll have a massive advantage over people who don't use Ahrefs, or who don't understand the importance of the traffic metric.

To me, this was a major lightbulb moment.

And Tim's work isn't yet done. I spoke to many people after the conference and asked what their biggest takeaway was, and nobody thought of the traffic metric revelation.

Even SEO's can't see the woods from the trees sometimes, and that's what compelled me to write this post.

The sooner you take advantage of this, the better.

What About The Keyword Difficulty Score?

You should not really be relying on any particular tool to tell you how difficult a keyword is. You can use the score as a brief filter for finding out which keywords might be impossible and which are worth investigating deeper, but that's about it. To really know which keywords are easier or harder, you have to learn how to analyze a SERP, something which is beyond the scope of this article.

Blog Post: Combine keyword research tools with manual user experience.

How We'll Be Using This Information Moving Forward

We were pretty much ready to move away from SECockpit before I saw Tim speak, but now it's a done deal. As of the next couple of weeks, all HPD websites, articles, and keyword packs will be researched using Ahrefs.

We'll use traffic data rather than search volume to do our niche validation and keyword research, and we'll build our sites and content silos around these topics, which will give ourselves and our customers a massive advantage.

I used to feel that the only thing differentiating keyword tools was their Keyword Difficulty metrics, which meant that there wasn't really anything accurate between them, since KD scores are usually informative, but far from conclusive.

Now I'm really excited to dig into Ahrefs keyword explorer knowing that I'm using a far superior tool.

  • Hey Dominic,

    I remember 3 years ago I got a chance to get an inside look at Ahrefs and back then I was pleasantly surprised. I understand where your coming from and thinking they all scrape the Google Planner tool. I figured they were probably all about the same too when I checked most of the big keyword research tools out for myself.

    I was married to Moz for about 2 years, though. But, Ahrefs actually does show a lot more links than Moz. I could be wrong on this but I believe Ahrefs has been around since 2010. It’s nice to see that newer tools are actually pulling more accurate data.

    But, have you checked out the content explorer tool with Ahrefs? I put so much emphasis on producing quality content and less on building any backlinks. For years I used to think it was all about having backlinks, but just focussing on content will attract backlinks naturally.

    Did have a question for you though? How do you feel Jaaxy compares with Ahrefs in terms of finding keywords that you can “actually” rank for on Google?