Revealing the Secrets of Keyword Data
Keyword search volume is probably the most frequently used metric in SEO. Digital marketers rely on it for making important strategic decisions and prioritizing their content calendars. However all too often people take this metric at its face value without understanding where it comes from and how accurate it is. And this is exactly what we’re going to discuss with Tim Soulo, the Chief Marketing Officer at Ahrefs, one of the leading SEO tools that provides keyword data.
Scheduled for 22 March 2022 at 11 H CET (Paris)
The event is 100% free:
Organised by Kalicube in partnership with Wordlift and Digital Olympus.
Part of the Kalicube Tuesdays series.
Listen to the podcast episode >>
Helpful Resources About His Work:
Transcript:
[00:05:17] Jason Barnard: Welcome. Brilliant! Wonderful! I always start off with the Brand SERP. This is my kind of little thing, my rabbit hole that I go down and hey, we’re looking at the Ahrefs’ Brand SERP in London. That looks brilliant. What I love about that is you’ve got your Rich Sitelinks. You’ve got the really, really nice top stories. That means you’re producing a lot of content that Google sees that that content that you’re producing on a very regular basis is incredibly valuable to your audience because it shows on your Brand SERP what it feels to be valuable, helpful, and relevant to your audience.
[00:05:47] And that for me is an amazing sign of how good your content strategy is, Tim and your team. So, all credit to them for that. That’s amazing! And the other aspect is the entities that Google’s understanding. Looking at the bottom of the SERP, we can see Ahrefs as an entity and the understanding that Google has of its relationship with all of these other entities, and that’s a boatload. And I find that incredibly interesting from my perspective, entity SEO, entity optimisation. And as Koray Gübür says entity identity are now one big aspect of SEO that we need to take care of. I’m actually working with you guys, Ahrefs, a little bit on that and it’s really enjoyable. You guys are really smart and we’re doing some great work. None of that is thanks to me, of course, it was done before I even turned up on the scene. But that’s a great Brand SERP, Tim.
[00:06:42] Tim Soulo: Thank you. We’ve actually had some troubles with our Knowledge Panel because apparently people who work or the editors of Wikipedia aren’t particularly fond of SEO people. I believe that’s because SEO people like to spam their website with their stuff to be able to get links from Wikipedia, and so it seems that some editors got allergic to any mentions of SEO tools and the SEO industry. So, we had some troubles with that, but hopefully we’ll eventually find a way to display a nice-looking Knowledge Panel in our SERP.
[00:07:25] Jason Barnard: Yeah, well, that’s exactly where I’m helping you. And it is a problem that a lot of companies are having especially in the SEO community, but also in other industries, creating a Wikipedia page can be very delicate, very tricky. You’re not supposed to do it yourself, but without Wikipedia, it’s certainly possible and definitely, definitely something I’m very keen on is to start building Google’s knowledge, educating this child that is Google without Wikipedia, without even Wikidata, we can do it. It’s a child that wants to understand, it just needs us as responsible adults to explain clearly and precisely who we are, what we do and who our audience is, but enough about my pet topic and onto your pet topic.
[00:08:06] Tim Soulo: I’m just hoping that this child won’t grow into an angry teenager.
[00:08:11] Jason Barnard: Ohh, that’s a very good point, yeah. So, I hadn’t taken the child analogy that far. But now you’re going to give me sleepless nights. That was the worst thing you could possibly say. Now I’m frightened. Brilliant! Let’s get onto keywords because you know lots about that and I don’t. Now, when you’re saying revealing the secret of keyword data, you’ve got boatloads of keywords in Ahrefs. How many billions of keywords?
[00:08:34] Tim Soulo: I don’t even remember the number by heart because it’s so uncomprehensible that you cannot connect it to anything else in your brain and make it meaningful. It’s like, I don’t know.
[00:08:45] Jason Barnard: Right.
[00:08:46] Tim Soulo: I think in the U.S., we have 4 billion but is it a lot? Is it little? You don’t know. Probably compared to how many different search queries people are entering in Google, that’s little. But are we covering all of the most important search queries? I would think that we are, so that’s basically what matters.
[00:09:11] Jason Barnard: Right. Yeah, no, a 100% of a new thing. I mean the very, very long tail, it’s impossible to actually get your hands on it. Google isn’t sharing the data. And I mean, I suppose when your users, your clients, enter new keywords, that’s giving you an idea of where people are going, as opposed to where Google thinks they’re going. Is that helpful?
[00:09:28] Tim Soulo: No. We don’t work with any user input. There are lots of conspiracy theories about Ahrefs looking at what people are doing in the platform and then somehow changing the platform based on that input, but no we are not doing this because it creates a huge loophole that people would be exploiting. So, we don’t want to do this. We don’t want people to be able to influence the data that we have just by entering it into Ahrefs.
[00:10:03] Jason Barnard: Oh right, I didn’t even know there was a conspiracy theory, I do apologize. I just meant that when I add a keyword to Ahrefs, you then track it. Therefore, I thought it got added to the whole pool of keywords.
[00:10:13] Tim Soulo: Ahh, no. I saw that when you’re putting a keyword that is not in our database, we will then edit our database. So, no, we don’t use any input from people. The same way as if you are inputting some domain that we haven’t crawled yet, it doesn’t mean that we’ll somehow prioritise this domain for crawling because one of our users have entered it to Ahrefs. This is a slippery slope.
[00:10:43] Jason Barnard: Right. I get the feeling now, it’s feels like I’m talking to somebody at Google. It’s similar as what’s happening internally in Ahrefs and people are actually very interested in that and trying to manipulate it or potentially trying to manipulate it. I’d never looked at it that way.
[00:10:57] Tim Soulo: Yeah. Yeah. People like to manipulate our metrics. For example, the domain rating metric, which indicates the I call it link popularity over domain. So, how many other domains from all around the world web linking to you. It is then indicated in a simple two-digit number. Well, sometimes I don’t think anyone has domain rating of a 100, so yeah, two-digit number. And yeah, people are looking for ways to game it because there are lots of links sellers and they’re selling their links on the basis of how high the DRO linking website is. So, they are very much motivated to grow their DR with any loopholes and then sell links at a higher price.
[00:11:42] Jason Barnard: Wow. I feel so naive when you tell me things like that because I just hadn’t thought about it, hadn’t occurred to me, but in fact, there’s all those nefarious kind of tricks that people play using different metrics and you actually manage one of those metrics. So, it becomes a battle for you too. Right, sorry. But back onto the topic, I mean, that was incredibly interesting and I’ve just learned lots of things I didn’t know before, but you were talking about keyword search volume. We all use that as a metric. Is that really the best metric?
[00:12:13] Tim Soulo: Great question. So, we’ve been dealing with that for many, many years ever since we started introducing different keyword research tools. And the common complain or criticism that we were hearing is that our numbers don’t match those of Google Keyword Planner.
[00:12:34] Not that Google Keyword Planner didn’t have its own troubles and I believe the first person to criticise the data from Google Keyword Planner was Ralph Jones from us who is unfortunately no longer with us, but I will still be giving him credit for all his SEO work because that he was an absolutely brilliant person.
[00:13:00] So, Google Keyword Planner isn’t without a sin, but somehow the general public in the SEO community was thinking that the numbers that come from Google Keyword Planner are the single source of truth. But then, we also have Google Search Console and those SEO professionals who are more attentive and who had the chance to compare the number of impressions that they were getting in their Google Search Console if they were consistently ranking at the top, with the data that was coming that Google Keyword Planner was showing them, they already noticed that there’s quite a bit of discrepancy. And the more hardcore it shows, knew for quite a while that search volume that you’re getting from Google Keyword Planner or pretty much any SEO tool is not a particularly precise metric.
[00:14:03] Jason Barnard: Sorry. No, it’s just a question about keyword planning because it always struck me that the Keyword Planner is there to sell Google Ads. So, Google have got every motivation to hide data slightly, to round it up to a nice big number when it should be 400 they’ll round up to 500. I mean, I don’t know what they do, but the motivation of Keyword Planner is to sell Google Ads, so that isn’t necessarily compatible with actually reporting on search volume.
[00:14:34] Tim Soulo: Well, how long have you been in SEO, Jason?
[00:14:37] Jason Barnard: Oh, I started 25 years ago, but I’m old.
[00:14:42] Tim Soulo: So, I’m not surprised that you are thinking of such things because for a person who is just I know two or three years in, it doesn’t really cross their mind that this point that Google Keyword Planner is a tool for advertisers and they might be adjusting, let’s use the word adjusting, their search volume data in a way that would benefit them.
[00:15:07] And then, they look at the search volume numbers that they see in Ahrefs or pretty much any other third party tool that reports this data. And they’re unhappy because those numbers don’t necessarily match. But there are reasons to that. So, up until recently, we couldn’t really create any compelling research to show people all the problems or like how far off Google Keyword Planner is.
[00:15:39] But since we introduced what free layer of Ahrefs, which is called Ahrefs Webmaster tools, where you can get all data for your own website as long as you can verify it with a number of ways. And one of these numbers is to connect via Google Search Console. You can also upload a file, you can change your DNS, so basically all the common ways to verify the ownership of your website but one of them is Google Search Console which a lot of people are using because we also started displaying Google Search Console data in Ahrefs if you connect it.
[00:16:12] Jason Barnard: Right.
[00:16:13] Tim Soulo: So, we started getting lots of Google Search Console data of our customers and then we realised, hmm, we can do a research and we can compare the search volumes or impressions from Google Search Console for a number of keywords. With the data that we see in Google Keyword Planner and see how those two data sources compared to each other.
[00:16:40] So, what we figured is that in the majority of cases, I don’t remember the exact numbers, but maybe we can later link to the research study that we published on Ahrefs book. But in the majority of cases, Google Keyword Planner was overestimating the search volumes compared to what you get in Google Search Console. And the actual reason for that is pretty well-known again to seasoned SEOs is because they group keywords with similar meaning together and they display aggregate search volume for the entire group. Again, this is because it’s a tool for advertisers. So, if you want to advertise for a particular keyword, for example, you don’t want to bother about singular and plural versions of that keyword like cat food, cats food, cat foods, cats foods, I don’t know, something like this, food for cats, To Google Keyword Planner, those things are the same now, but if you are an SEO professional, you do want to know the most common way how people search for something because then you want to use that in your copy to speak to people in their own words.
[00:17:52] Jason Barnard: Yeah. No, but that’s incredibly important. I think a lot of people miss that boat, as it were is that they think, oh, I need to match exactly that phrase because that’s what Google likes, but it doesn’t actually carry. What does matter is that you’re speaking the language that your users are using not just the language of the English and French or Ukrainian, but speaking using the vocabulary and the turns of phrase that people actually use.
[00:18:19] And I think kind of focusing on the user, which is what Google says all the time, where people get very angry with John Mueller for saying that over and over again, it does sound kind of trite and banal, but it’s actually what you need to do. Sorry, that was my little rant there.
[00:18:34] Tim Soulo: No, no problem. Absolutely. The focus should be on the user and they think if you want to future proof your website or future proof your marketing or SEO strategy, you should definitely have a user-first approach because Google is absolutely optimising towards what users would like. So, if they are not there yet, in three years, five years, they will be there, or at least they will get much closer. So, if you’re planning for your website to exist for the next three to five years, you should be optimizing for people because if Google is not realizing that you’re optimizing for people today, they will understand it in a year or in two years, and you will blow your competition out of the water.
[00:19:26] Jason Barnard: Right. Yeah. It comes out. If you’ve got any kind of long-term business strategy aim for where Google’s going, whilst making hay while the sun shines today as much as you can. But that’s a bit of it and the other thing is focusing on the users. It’s just occurred to me, John Mueller talks about Pull Queries, which is Brand Queries and he says, people aren’t focusing enough on that and you really need to start focusing on that, which is great for me. But when I look at Brand SERPs, it’s the only thing I do is think why is this useful, helpful, and relevant to the user, or at least why does Google thinks it’s relevant and helpful and valuable to your audience who are a subset of its users?
[00:20:03] And I’m writing an article right now for Wix about Rich Sitelinks and they seem so boring, those big sitelinks that we showed everyone at Ahrefs. But there’s so many insights you can get from that is what is Google thinking? How is Google approaching this? And what are my audience actually looking for? Sorry. Once again, you’re making me rant.
[00:20:23] Tim Soulo: No, not a problem. It’s very on point and it’s funny you mentioned Wix because it gives me a great bridge to another kind of fallacy that people have with search volume. A lot of people focus too much on the search volume of individual keyword but like we just discussed, there are tons of different ways to search in Google for the same thing. And these days, Google is smart enough to understand that those searches mean the same thing and rank basically almost the same set of pages for all the searches. So, what you should be looking at is the total traffic potential over a given topic rather than the search volume of an individual keyword. That is just one of the many, many ways to search for this topic. For that reason, we believe in that concept so much that we productised it. We created the metric that is called traffic potential. Basically it’s very simple. We don’t like to overthink our metrics because then it is harder for our users to understand them.
[00:21:30] So, what traffic potential does, it shows you the traffic to the number one ranking page for you in keyword. So, not only you see how many people search for cat food, specifically this keyword cat food per month, but you can also see how many clicks, estimated clicks, are landing on the top ranking page for cat food because it also ranks for lots of other similar queries. So, that’s the thing and why I mentioned Wix is because Wix is such a strong brand. You would be probably the best person to talk about this, but they rank number one for website builder with their homepage. Which is why some people are saying that our traffic potential metric is bad because when they put in website builder, they get the search traffic potential of Wix homepage which is huge.
[00:22:23] Jason Barnard: Right.
[00:22:24] Tim Soulo: What we explain back to them is that this is what it happens. To rank for website builder, it takes creating a brand. So, you’re not competing with some random page that ranks for website builder, and that is to optimise for website builder, you’re competing with an actual brand. So, it is insightful in itself that you no longer can rank for website builder by optimising for it because Wix now occupies the number one position for the term.
[00:22:53] Jason Barnard: That’s a really interesting point from the perspective of saying who am I actually competing with? And looking at website builder, you kind of say, right, well, I can optimise my page, and then you look at that first page, I’ve got no idea what’s on there, but certainly it’s going to be the really big hitters and a lot of people I talked to immediately come in with that kind of queries. I mean we’ve got absolutely no hope. You need to build from something else that’s more realistic for a smaller brand like yourself. And a lot of people they are looking at that keyword volume, they see website builder, which is exactly your point and they go super, super overboard about it and just focus on that all the time.
[00:23:30] I had a client like that with a French keyword but we did actually manage to end up ranking for that very short head keyword but it took us four years and we had to build the brand as you say, which is brilliant. So, what should we be looking at beyond obviously volume? I liked the idea of the potential of a given page. What else should we be looking at?
[00:23:49] Tim Soulo: Another reason, basically, you should be looking at the potential of the total traffic to top ranking pages and again, adequately predict your chances of where you’re going to rank because you’re definitely not ranking number one for website builder unless you build a company as big as Wix. So, that’s the thing and another reason why it is important to actually look at the traffic of the top ranking pages other than the search volume of individual keyword is because Google is stealing clicks from us. The most straightforward way on how they still clicks is with ads. They show freaking four ads on top of their getting search results. And if you look up some search queries on mobile, it gets absolutely ludicrous because those ads might also contain images, they might also contain sitelinks. So, you have to scroll a couple of times with your thumb until you reach the first ranking search result. So, even if you rank number one for something, it doesn’t mean you’re getting loads of traffic because you have ads above it. And then, Google has those SERP features which is why at the beginning of our conversation, I mentioned that I hope Google won’t become an angry teenager the more we educate him because what bothers me is that the more Google understands how to read, process, curate and display information, the less they need to send people to the actual websites. I don’t understand why the SEO, excuse me, I’ll finish my thought.
[00:25:39] I don’t really share the excitement of the SEO community with the artificial intelligence tools that help you create content, those GPT-3 and such because if you are going to generate your content with the robot, then what is the motivation for Google to send people to your website while they can use the same, if not better system, to generate that content right in their search results page?
[00:26:12] So, you’re actually putting yourself at a disadvantage by relying on those tools and by being happy that you can just generate content. I am very afraid of that. I don’t like that this is happening. I don’t really like that Google is educating themselves. Because again, if you look at many content marketers, what they’re doing is that they search for something. They open the few first pages of the search results. They read that content and they rewrite it into a different article. It’s the same content. They just read it somewhere else. They rewrote it without adding anything fresh, anything new, they’re not subject matter experts. And then, with the backlinks coming from, I don’t know, PBNs or like other people who are willing to link to yours as long as you pay the money. SEO professionals put this page at the top of the search results and what you get is a mediocre content that is made from other content that is ranking at the top of Google. Why would Google want that to happen? Why won’t they use their own GPT-3 or something to, again, aggregate the content from those top ranking pages and rank their own page at number one, or even display that in the Google search results. So, yeah, I’m very unhappy about that trend.
[00:27:32] Jason Barnard: Right. Okay. That was your rant. So, we’ve both had rants and that was a delightful rant. I do like it because it’s philosophical and speaks a lot of truth. I mean from that perspective while you were talking about that, I was thinking about multiple things.
[00:27:45] One of which is in the People Also Ask, Google is increasingly just putting facts there with no link especially on Brand SERPs, once again, and Lily Ray was pointing out that if you search for best of something or even health magazines, it will give you a big block of entity filter pills. And so, it’s actually now recommending, so those best of articles are actually going to be more and more problematic or less and less valuable to people because Google say, right, this is what I’ve understood, so we then end up in a situation where if you’ve educated Google about who you are, what you do, and who your audience is, then you can get in there. But then we’re ending up in this situation where people are researching the entire topic on Google and our only chance of the visit is when they’re pretty close to bottom-of-funnel.
[00:28:29] Tim Soulo: Mm-hmm.
[00:28:30] Jason Barnard: Don’t know if that was actually a question. It was just a statement of things that I was thinking while you were talking. But your top stories, they’re the top of your Brand SERP, I think that’s incredibly interesting because that’s something that Google cannot replicate. Your insight, your expertise in stories or articles that you’re writing regularly. There’s Joshua who does a lot of that and the guy who does your video content as well whose name I’ve forgotten I’m afraid. Isn’t he Soh? Sam Oh is it?
[00:28:56] Tim Soulo: Sam Oh.
[00:28:57] Jason Barnard: Yeah. What a great name. How could I forget that?
[00:29:00] Tim Soulo: Hahahahaha. Yeah, both guys are incredible. But in terms of, I’m not an expert in Brand SERPs, or even SERPs. You are the much more experienced here, so I might be wrong in my assumption. But I’m thinking that the reason why Google displays those articles in our Brand SERP is because they think we’re some kind of a news outlet in our industry which we are not. So, there are different kinds of blogs. For example, I’m sure like most people know about Search Engine Land, Search Engine Journal, Search Engine Watch. These are the blogs that are covering the latest developments or the latest news in the SEO industry. So, if you want to know what were the latest statements from Google or what were the latest releases from Google Search Console or Google Analytics, you would go and visit those websites because they always cover the freshest stuff.
[00:30:04] In the case of Ahrefs, we have what I call a resource blog. So, we don’t necessarily write about the hottest, latest and trendiest topics. We just research what people in our industry are searching for and we create resources for them to learn from. So, probably because of our publishing pace which is quite consistent and rather fast, Google might think that we are one of those news blogs, so this is why they feel important to display those latest articles in the Brand SERP.
[00:30:40] I’m not sure if it’s true, but what I’m alluding to is that our content is educational content rather than news content.
[00:30:50] Jason Barnard: Right.
[00:30:51] Tim Soulo: Google cannot really be a source of news content because someone has to take it from the real world and put it into words. In terms of educational content, the amount of that content is growing all the time. So, take any topic whether it is in marketing, in photography, in parenting. There’s lots and lots of content that’s already written. And most people probably there’s Pareto principle at work here, 80% of people are searching for the same basic things because as you go more advanced and start looking for advanced topic, there would be less of these people.
[00:31:31] So, what is stopping Google with all those GPT-3 and such to create those resources like Wikipedia. But Wikipedia is created by humans while Google can just create their own curated Wikipedia based on what has already been written on the topic. So, yeah, if you will offer something brand new, some kind of news-based thing, then you’ll be able to get visitors to your website. Otherwise, if you’re talking about the same topic that a hundred different websites have already covered, why wouldn’t Google just create kind of sort of a Wikipedia page based on what they’ve learned from those hundred pages. So, this is what I’m ranting about. This is what I’m afraid of.
[00:32:19] Jason Barnard: Yeah and I mean, it’s already happening. In Knowledge Panels, we’re seeing sometimes a little snippet that’s actually just written by Google using GPT-3 or whatever it is, no longer Wikipedia or IMDb or even my own website. I mean, from a Knowledge Panel perspective, I’m advising clients to say you need to get that description from your own website. But at a certain point in time, Google’s just going to write it itself, as you say. And from my perspective, that also means you need to make sure that it gets it now, because then it’s more likely to write what you want it to write as opposed to what other people are writing. But that’s a totally different topic. What I’d like to do is give you an opportunity to say anything that we didn’t talk about because we rather went off topic there and it kind of moved around. Is there anything that I didn’t ask that you really wanted to share about the secrets of keyword data other than don’t focus on search volume?
[00:33:09] Tim Soulo: I think that’s actually the most important takeaway. If we’ll get into details, the right format for that would be actually the research article that I’ve published because you need to see the graphs, you need to see the exact numbers, which I don’t remember by heart, but there are a couple of takeaways. So, the first takeaway is that don’t think of Google Keyword Planner as having accurate search volume numbers. First of all, don’t expect third party tools like Ahrefs to mimic the numbers from Google Keyword Planner because our goal is, at Ahrefs, I’ll speak for ourselves, our goal is to use the other data that we have to refine those numbers and give you a slightly more accurate picture.
[00:33:55] Like we said, Google Keyword Planner is for advertisers, we are a tool for SEO professionals. So, we are trying to solve the small issues that Google Keyword Planner has, fix them, and show SEO professionals the data that they are looking for. So, yeah, don’t expect a tool like Ahrefs to give you the same data as Google Keyword Planner because if you need data from Google Keyword Planner, please go use Google Keyword Planner. I think that’s quite fair.
[00:34:21] The second thing is that search volume of an individual keyword is not always indicative of how much search traffic you’re going to get. Because it all depends on the whole kind of cloud of search queries that mean the same thing. And sometimes that we can look at it like an iceberg. So, sometimes there’s a super popular way to search for something and underneath there’s a little bit more of similar searches. The topic is very narrow and it is very defined.
[00:34:58] For example, a Squeeze Page. There’s a marketing concept called a Squeeze Page. There aren’t too many ways to search for a Squeeze Page because it’s such a narrow topic. The amount, the cloud of related searches is very small, but if we take a topic of how to get more traffic to a website, 10 different people would phrase it differently all the time.
[00:35:21] So, even though and you cannot even pinpoint the most popular search query because there are many of the most popular search queries. So, if you’re looking just at one of them, you’re misleading yourself as to the total search traffic potential of the topic. So, it is very important to actually review the top ranking search results. And then, the traffic that is coming to the top ranking pages as estimated by Ahrefs or any other similar tool accounts for all those things that we just discussed, the ads, the Knowledge Panel, the People Also Ask Boxes, we account for those things because they still clicks away and we adjust the traffic estimation based on those things.
[00:36:03] So, even though the keyword or even the overarching topic would be incredibly popular, if there are too many ads, too many SERP features, and you won’t even rank number one, you’ll rank somewhere in top five, you won’t be getting too much traffic. So, these are the things that are important to know and understand because at this way, you’ll be able to adequately prioritize your list of topics that you are going to go after.
[00:36:35] Jason Barnard: Brilliant. So, we can’t hit and guess anymore. We actually have to use data and that goes through platforms like Ahrefs and other search platforms. Thank you so much, Tim. That was absolutely brilliant. We’re just going to present next week. Now passing the baton, next week is going to be Elizabeth Marsten, Retail Media in Search, that’s going to be absolutely delightful, e-commerce, we don’t cover that enough. Incredibly insightful. She’s incredibly smart. And it’s going to be really, really interesting. Tim, could you pass the baton?
[00:37:04] Tim Soulo: Elizabeth, I’m passing the baton to you and I’m hoping that you would have as great of a conversation with Jason as I just did.
[00:37:12] Jason Barnard: Oh, you’re such a charmer. Thank you so much for that charming comment. Oh, I’m delighted! A quick goodbye to you and the show. Thank you to Tim.
[00:37:25] Tim Soulo: Thank you, Jason.
[00:37:26] Jason Barnard: And thank you everyone for watching. See you next week.
[00:37:28] Kalicube. It’s all about your Brand SERP.
