For many years now, Moz's customers and so, so many of my friends and colleagues in the SEO world have had one big feature request from our toolset: "GIVE ME KEYWORDS BY SITE!"
Today, we're answering that long-standing request with that precise data inside Keyword Explorer:
This data is likely familiar to folks who've used tools like SEMRush, KeywordSpy, Spyfu, or others, and we have a few areas we think are stronger than these competitors, and a few known areas of weakness (I'll get to both in a minute). For those who aren't familiar with this type of data, it offers a few big, valuable solutions for marketers and SEOs of all kinds. You can:
- Get a picture of how many (and which) keywords your site is currently ranking for, in which positions, even if you haven't been directly rank-tracking.
- See which keywords your competitors rank for as well, giving you new potential keyword targets.
- Run comparisons to see how many keywords any given set of websites share rankings for, or hold exclusively.
- Discover new keyword opportunities at the intersection of your own site's rankings with others, or the intersection of multiple sites in your space.
- Order keywords any site ranks for by volume, by ranking position, or by difficulty
- Build lists or add to your keyword lists right from the chart showing a site's ranking keywords
- Choose to see keywords by root domain (e.g. *.redfin.com including all subdomains), subdomain (e.g. just "www.redfin.com" or just "press.redfin.com"), or URL (e.g. just "https://www.redfin.com/blog/2017/10/migration-patterns-show-more-people-leaving-politically-blue-counties.html")
- Export any list of ranking keywords to a CSV, along with the columns of volume, difficulty, and ranking data
Which keywords do you rank for?
My top favorite features in this new release are:
#1 - The clear, useful comparison data between sites or pages
Comparing the volume of a site's ranking keywords is a really powerful way to show how, even when there's a strong site in a space (like Sleepopolis in the mattress reviews world), they are often losing out in the mid-long tail of rankings, possibly because they haven't targeted the quantity of keywords that their competitors have.
This type of crystal-clear interface (powerful enough to be used by experts, but easily understandable to anyone) really impressed me when I saw it. Bravo to Moz's UI folks for nailing it.
#2 - The killer Venn diagram showing keyword overlaps
Aww yeah! I love this interactive venn diagram of the ranking keywords, and the ability to see the quantity of keywords for each intersection at a glance. I know I'll be including screenshots like this in a lot of the analyses I do for friends, startups, and non-profits I help with SEO.
#3 - The accuracy & recency of the ranking, volume, & difficulty data
As you'll see in the comparison below, Moz's keyword universe is technically smaller than some others. But I love the trustworthiness of the data in this tool. We refresh not only rankings, but keyword volume data multiple times every month (no dig on competitors, but when volume or rankings data is out of date, it's incredibly frustrating, and lessens the tool's value for me). That means I can use and rely on the metrics and the keyword list — when I go to verify manually, the numbers and the rankings match. That's huge.
Caveat: Any rankings that are personalized or geo-biased tend to have some ranking position changes or differences. If you're doing a lot of geographically sensitive rankings research, it's still best to use a rank tracking solution like the one in Moz Pro Campaigns (or, at an enterprise level, a tool like STAT).
How does Moz's keyword universe stack up to the competition? We're certainly the newest player in this particular space, but we have some advantages over the other players (and, to be fair, some drawbacks too). Moz's Russ Jones put together this data to help compare:
Click the image for a larger version
Obviously, we've made the decision to be generally smaller, but fresher, than most of our competitors. We do this because:
- A) We believe the most-trafficked keywords matter more when comparing the overlaps than getting too far into the long tail (this is particularly important because once you get into the longer tail of search demand, an unevenness in keyword representation is nearly unavoidable and can be very misleading)
- B) Accuracy matters a lot with these types of analyses, and keyword rankings data that's more than 3–4 weeks out of date can create false impressions. It's also very tough to do useful comparisons when some keyword rankings have been recently refreshed and others are weeks or months behind.
- C) We chose an evolving corpus that uses clickstream-fed data from Jumpshot to cycle in popular keywords and cycle out others that have lost popularity. In this fashion, we feel we can provide the truest, most representational form of the keyword universe being used by US searchers right now.
Over time, we hope to grow our corpus (so long as we can maintain accuracy and freshness, which provide the advantages above), and extend to other geographies as well.
If you're a Moz Pro subscriber and haven't tried out this feature yet, give it a spin. To explore keywords by site, simply enter a root domain, subdomain, or exact page into the universal search bar in Keyword Explorer. Use the drop if you need to modify your search (for example, researching a root domain as a keyword).
Find out which keywords you rank for
There's immense value to be had here, and a wealth of powerful, accurate, timely rankings data that can help boost your SEO targeting and competitive research efforts. I'm looking forward to your comments, questions, and feedback!
FINALLY!! *breathes sigh of relief* *dies*
Many thanks to the team for months of work on the back-end of this. The Keyword Explorer features are a first step as we explore how to use this incredibly valuable data in other products and features.
What I love for now, though, is the workflow aspect. People are always asking "How do I know everything I rank for?" -- this may not be quite everything, but it's a ton, and Keyword Explorer makes it easy to push and pull that ton and move it directly into lists and campaigns for actionable insights.
On top of that, the frequency of the refresh is huge, IMHO. There is no need to export and filter out crap that is 2, 3 or even 6 months old. It is all within the last 2 weeks or so. This was a pretty big pet-peeve of mine with other tools.
Hey Rand,
A few things about Ahrefs are not quite right:
1) Ahrefs > "Full Corpus Refresh Rate" > 12 months
Actually, we update our Full Corpus in ~3 months.
The frequency of updates also depends on the search volume of each keyword. Thus popular keywords are updated every day, while long tails may take a few months to update.
2) Ahrefs > "Evolving Corpus" > N/A
Actually, our corpus works somewhat the same way you described yours:
> "We chose an evolving corpus that uses clickstream-fed data from Jumpshot to cycle in popular keywords and cycle out others that have lost popularity."
It's easy to compare our "Evolving corpuses" by looking at the sales page of the recently released iPhone X ( https://www.apple.com/sg/iphone-x/ )
Screenshot: https://imgur.com/a/dccQa
Hey Tim, thanks for the response.
I hope we can clear up the confusion here. If your entire corpus is re-crawled every 3 months, then why are there year old results? Here is a screenshot of the current Moz.com is a screenshot of the current Moz.com results in Ahrefs as of today which any of your customers can recreate. When I refer to "full refresh", I mean "what is the oldest result you will find in the index". Perhaps we are just talking past one another?
As for the evolving corpus, thanks for answering that question. I didn't get an answer to that via support on your site, but I had a intuition that you might use the clickstream data to evolve your corpus too. We will get that fixed in the post ASAP.
Hi Rand, Tim, Russ,
I think SEMrush also work on a similar refresh model where keywords with higher search volume are updated daily and keywords with lower search volume are deprioritized - but they will all receive an update within a certain time frame.
Rand, I like the two week data refresh because it's good to know the numbers will always be fairly recent and reliable.
However, I think it also makes sense to prioritize keywords with high search volume and have their data updated more frequently.
Is Moz updating high volume keywords more frequently within the two week full refresh? Do you think this is required?
I'm also wondering if Moz is working on adding site/page traffic estimates?
While seeing the number of keywords that a site/page is ranking for is helpful, I find it much more useful to see traffic estimates from those keywords.
Is this something else that Moz is working on? Or does dealing with keyword volume buckets make traffic estimates too difficult?
Finally, like others have mentioned, I'd love this tool to be expanded for use in Australia! :)
Cheers,
David
> "If your entire corpus is re-crawled every 3 months, then why are there year old results?"
That is obviously a bug with some "stuck data."
We operate a corpus of keywords that is 10X bigger than yours (including non US), and at this scale it is pretty much inevitable.
The % of this "stuck data" however is very low in relation to the size of our keyword corpus. And, as you can tell, only low-volume keywords are affected, which doesn't change the grand scheme of things much.
Let me pull a visual for that, for more impact: https://imgur.com/a/s158V
Oh, and by the way. This "stuck data" means that old positions weren't removed when new positions were added, so we have both. See here: https://imgur.com/a/Ks5Wa
We already applied the fix and it does not happen with the new data, just need some time to clean old stuff.
All in all, we regularly review the performance of our keyword corpus, resolve these kinds of discrepancies and patch our system.
Nice update to Moz pro. I like the Venn diagram for competitive analysis. It's nice to see how different companies compare.
However, unfortunately Moz is targeting the American audience again and forgetting users around the world. As at the moment even keyword explorer isn't good enough for an Australian market.
This tool brings up very few keywords for smaller local businesses and I'm not confident with the accuracy of the keyword rankings for larger companies. Already searched a few and the rankings were wrong (it might be better in the US). Hopefully they get better at tracking volumes for non-american keywords soon and look into keyword explorer at the same time.
Anyone else used it around the world and have any feedback?
Hi Sally,
You are right, we haven't launched a product for the AU market yet. We are definitely working on it though!
Russ
Hi Sally
That's why I switched to other tools when doing keyword and link research. I think Ahrefs has the biggest Australian keyword database and link index.
Yeh definately sticking with Ahrefs and SEMrush for the moment!
100% right Sally. Moz's keyword universe is very US-centric. My advice would be to check out SEMRush, Ahrefs, and Spyfu, all of whom have better coverage in Australia (and many other countries) than we do. Hopefully, long-term, we'll get there as well.
Oops, looked like I just repeated Sally's question. Thanks for the answer, Rand, that covers what I was asking. FWIW I love Moz for your heart and attitude and every time I use SEMRush, Ahrefs etc I wish I were using Moz. So I am looking forward to you improving your AU coverage so I don't have to use them :)
Hi Rand (and the team),
it’s great that you accumulated the information about all tools in one table, it’s always helpful for users, but some of the data about SEMrush toolkit’s capabilities is incorrect, we’ll appreciate it if you update it:
- The number of keywords we have in our Analytics reports is 120 millions for US. Also, it looks confusing to me that for some tools including SEMrush and Moz you state the keywords for US and for some you use full database and some other countries. In my opinion, this is misleading.
- On top of the analytics part, we provide 500 million keywords for US in our Keyword Magic Tool (and we do have volume for them)
- Regarding the keyword updates, we update our database in a live mode. Depending on search volume we refresh top million of keywords on a daily basis, midrange volume keywords once in 2 weeks, and all the rest on a monthly basis. Due to the tech issues only 1 million of keywords out of 120 hasn’t been updated within 3 months, but will fix that asap.
- All the new keywords that haven’t been on Google before - we also add on a monthly basis.
- Regarding the volume data source, we use hybrid (Adwords + clickstream)
- And in the last column you do not specify the region, so in total we provide 3,6 billions keywords.
Feel free to reach out to our marketing team next time, we’d be happy to support your study and give the updated information.
Hi Olga,
Thank you for your response! We used the latest published numbers we could find, so we will update it. We do explicitly say that KeywordSpy and iSpionage numbers include multiple countries. We were unable to get a breakdown from them by country via their support team, so we went with what was published on their sites.
In the table provided, we state that you have 505,000,000 US keywords in your Keyword Magic tool. Is that not correct? Is it only 500 million?
So the 3 month old data in your system is due to a tech issue. This appears to have been the same problem with Ahrefs. We will happily update the table to reflect this. As a paying customer of SEMRush myself, I just went by what the oldest data was that showed in your system. I hope you can see how one would be misled to believing that SEMRush doesn't update every month when there is data in the tool itself that shows otherwise. At any rate, we will update the table accordingly.
In the last column we are discussing US only. Moz has 1.2 Billion US keywords with volume. Ahrefs has 1.7B, and as you stated earlier, SEMRush has 500M.
I will update the image with an asterisk to point out that a bug is responsible for the older data.
Hi Russ,
Thank you for the super quick response! The correct data for the last column would be 505,000,000 as you stated.For the Keyword Universe column it should be 120,000,000 for the US. For the Data source: Clickstream + AdWords. And for the evolving corpus it's monthly, we're adding all the new keywords that haven’t been on Google before on a monthly basis.
An update in the comparison would be really cool. And please feel free to nudge me whenever you need any data for any other comparison tables, happy to send it over!
Thanks MOZ Team!!
We are already Moz Pro subscriber and these added KE features are making our decision more worthwhile. I am loving the ways of data states showing there with diagrams.
This is HUGE. Thank you!!
Fantastic update Rand/Moz Team! Definitely something that I've been clamoring for quite some time and really digging the presentation of the data. (Already used some of the charts to "sell" a new project.)
My question (and it is a loaded one) is can I now drop one of my subscriptions. (Currently subscribe to SEMrush, Moz and Ahrefs and honestly, I'd like it to drop down to two subscriptions, instead of three. AKA--my employer will only pay for two.) If we assume I'm keeping Moz (which I am) what tool does everyone feel is next best to have in your pocket? (Love to hear more from Olga and Tim! And sorry in advance for a question this loaded!)
This is a great Tool and really got me excited! Tried it out for some international sites and was awesome. I can see many applications for this Tool. Hopefully we can get it for Portugal as well in the future (as at the moment it looks like the data volume from Portugal is to low, to get any substantial results). Maybe a suprise when you come visit us next month in Lisbon during the Websummit?
IMHO, one of most important factors here is who has the most accurate search volume; unless site you're researching is small and I need to find a lot of KW w/ only 10 searches a month. This is if all else is pretty equal (EG user experience or types or reports available). Most metrics sit downstream from KW volume: screw that up and everything is off kilter. I did a data study for Neil Patel about differences in these tools earlier in the year, but it was probably one of the most poorly thought out I’ve done (in terms of being actionable!) Very tough to say who is most accurate given that Moz v. ahrefs tend to have different approaches to this, for example. For sure, as others have noticed, ahrefs has much higher numbers over all vs. SEMrush. I haven't been using the Moz KW tool much but will now that they have "by domain"! Also, Russ has done tons of hand's on SEO, having been the CTO of an SEO agency for a long time, so I do think that's a big advantage vs. SEMrush’s team in terms of planning what will be most useful and accurate for users.
This is going to save me a lot of time. Thanks guys.
I've been a Moz pro subscriber and reader for years. Hands down one of the best updates I've seen (even with the hiccups). Great job folks!
Great update. I will definitely be checking this out and see how to use is alongside the SEMrush and ahrefs tools that I already use in my competitive analysis process.
The keyword overlap visualization is great, it would be cool to see some type of dashboard with a few more visualizations to differentiate a bit more from some of the other tools. We would also be able to use the visualizations in competitive analysis reports that we produce for clients.
Looking forward to testing this new tool and providing addtional feedback.
Legit upgrade Moz! I like to use Moz tools as much as possible, and am glad to hear the Keyword Explorer is now on par if not better than more expensive tools available in SEMrush, Ahrefs, etc. Well done Moz squad!
Basicall I would love the new feature but I am sorry to say that the keywords that appear searching with the root domain are either absolutley nonsense (nothing to do with the page at all and the pages are all very well keyword optimized) or I get "There are no results for your search". Not really a good user experience...
This looks really great - I love MOZ tools. Is it available for the UK ?
Thanks, Nigel
Hi Nigel! We are gonna expand into other countries over time. UK is pretty close to the top of that list.
Thanks so much for letting me know Jon!
Game changer! This will probably (definitely) feature in all my SEO audit reports for the next few years I'm guessing. Well done Mozzers!
This is fantastic - definitely made my day!
Question: is this using the US search results only? I'm dealing with some Australian businesses that I wouldn't expect to rank on Google in the US but would expect to rank in AU, and I'm seeing no results for them. And when I click "root domain" in the dropdown, the geography selector disappears.
Haven't launched an AU version yet but it is in the works!
Congrats Moz Team.
Looking for the * in Russ Jones 's table, can't seem to find it.
It is there now, along with a few updates relative to more info we got from the AHrefs team.
You did it: you kept me awake past midnight playing around with another new tool. <_<
Really great - but I hope that "other geographies" will come rather sooner than later; it's not really too much use as it is right now for Germany (or other European countries).
Moz is really really becoming the best Seo Tool one can use! I am testing this new tool and am so sight!
Great job!
Hi Rand,
Really looking forward to this added functionality! Looking good for US companies, however as it seems no data for the Netherlands (.nl) is available. When can we expect this? Is there any forecast?
Many thx in advance!
Extremely useful feature addition. I can see that you can also use it to track the keywords of your own website which are not in the top 10 yet for search queries but are within knocking distance.
However low volume search terms are not reflected.
Is there a way to export the full set of data to Excel for this? I tried to run an export of this awesome data for a site, and it only included the 50 visible keywords in the list? Am I missing something? Is this a feature you'll be adding in the future? Thanks for all of the hard work and for updating and improving on the toolset.
Export should provide all keywords. I just successfully exported several thousands. Are you logged in? You might want to start a support ticket if you are a logged in pro-subscriber and you aren't getting the full export.
For me the problem is slightly different, the export file is limited to the first 10,000 (of ca. 70,000) KWs am logged in when exporting the data. Thanks
Cool feature. Look forward to playing with this.
This looks like a great tool and definitely a useful addition. I'm excited to use it.
However my first test was with a client that we are currently running a MOZ Pro campaign. The problem is we're getting conflicting data.
Two points of clarification before I share the results though. This is how I understand the two "tools" work.
Here are the results.
If my understanding is correct in the first two points, then our keywords set should be equal or a subset of all the keywords we are ranking. Therefor how can Keyword explorer be showing less #1-3 keyword rankings than our keyword set?
Hi Brian - that's pretty much exactly what I'd expect from each tool. Keyword Explorer's dataset DOES NOT get your private rank tracking data and include it in our 40 million keyword set. We think that would be violating privacy and exposing risk. So, we constructed our keyword universe independent of individual campaign's rank tracking.
Thus, if you're tracking keywords in your private rankings inside Moz Pro that we didn't include in our public keyword universe, you should absolutely see more keywords ranking there than in the Keywords by Site inside KW Explorer.
Definitely a feature, not a bug :-)