Skip to Content

How To Adjust Your Keyword Research Strategy For The Google Helpful Content Update

54 min read

The latest content update which Google rolled out is the most significant one yet. To stay competitive you need to adjust your keyword research strategy. Join Dmitry from TopicRanker.com and Jeff from MarketMuse to find out how your keyword research strategy should change now that the new Google algorithm is in place.

Dmitry and Jeff focus on how to find keywords your website can easily rank for on Google due to specific problems and weaknesses in search results.

Click to view the entire conversation.

They dive deep into how each one of the following factors presents a specific problem in a SERP and how you can find keywords you can easily rank for due to the following weaknesses:

  • Lower DA sites ranking high on SERP
  • Title tags on SERP do not match search intent
  • Poor Readability score
  • Poor topic depth
  • Word count is under 1K words on SERP
  • Date is older than 8 months ago on SERP
  • Load time of the page is greater than 3 seconds

Show Notes

Here are some highlights from the conversation.

Recent Google Updates and Where The Puck Is Going

Dmitry’s take on these updates is that “muscling your way up with links is not going to be the way things are going to be done anymore. Like it’s going to be about how the content, like topical depth and the content itself, how truly comprehensive and helpful it is for people.” He relates an experiment he did with his new site, TopicRanker, where he managed to rank a page without links. The idea was to “throw up the best piece of content, promote it on social, promote it through email.”

Google’s Helpful Content Update & What to Do With Low-Quality Content

Dimitry feels that “you need to have your niche and that needs to be your topical depth and that’s what you’re really good at.” He believes that, when cleaning up a site, all pages must support the primary focus. “And if it doesn’t, maybe it’s time to part ways with it and let that go, because it just doesn’t, it doesn’t really serve you anymore. It used to be where you could have a blog about all sorts of things. But now I feel with this update really should be one main focus.”

To Prune or Not to Prune

Dmitry explains his process for making the decision of whether to keep or trash specific pages. Key to this is making sure that any content that is kept relates to the main focus of the site. He offers an example from JustReachOut where, if the page is not aligned with the content focus, they’ll most likely delete it.

Jeff points out that “there are so many different processes for evaluating.” He notes that Dmitry is “putting a lot of experience led decision making into that process.” Often marketers will sort descend based on traffic and delete what’s at the bottom. Jeff says that’s “like pulling teeth outta your jaw. There are pages that are really important to your topic authority that don’t get any traffic.”

How to Cater Your Strategy to Both Low and High Volume Keywords

A question from the live audience centered around targeting low volume keywords in a new strategy vs the high-volume keywords they previously targeted. Dmitry cautioned that topics need to be related regardless of their search volume. In addition to topical alignment, you need to understand whether you have the topical depth and authority to rank for it. Regardless of search volume, you should look for weaknesses and problems within the SERP that you can exploit.

Intent Mismatch Signals

Dmitry describes how TopicRanker uses some “AI that tries to extract meaning from the word that they put in, into the search.” They “extract meaning from the actual term” and then grade how it fits “with the meanings of the other other search results that are on the first page of Google.” Anything less than 80% match is considered a weakness. However, Dmitry cautions that “this is not an exact science.”

SERP Difficulty Signals

Dmitry explains that TopicRanker has some signals that it uses to evaluate the difficulty of a SERP, including lower Moz DA sites ranking high on the SERP, title tags on SERP don’t match the search intent, low readability score, shallow topic depth, low word count, older pages on the SERP, and they load slowly.

Featured Guest

Dmitry Dragilev

Founder & CEO, TopicRanker

LinkedIn Twitter

Resources

TopicRanker

JustReachOut

CriminallyProlific

Keyword Research and The Search Volume Illusion

Content Strategy Example: Combining Competitive Analysis, User Intent, and Topics

How & Why You Should Target Low Search Volume Keywords

How I Ranked #1 Without Any Backlinks – Hint: Google Helpful Update Helped!

Which Keywords Are Best To Target? [Step by Step Guide]

Transcript

Jeff Coyle: Hello, welcome to another MarketMuse content strategy webinar in our series. I’m Jeff Coyle, the co-founder and chief strategy officer of MarketMuse. And today we’re talking about something obviously near and dear to my heart if you watch any of these recordings. But we’re talking about keyword research and how to adjust your keyword research strategy for specifically for recent updates, most notably the helpful content update.

And I’ve got someone who is very qualified to speak about this has done probably as much of research on this specific update and historical updates as anyone. But before I do that, like that little teaser, I want to do a little housekeeping, ask us anything that you want. We will either get to it in line if it’s directly relevant to our discussion, or we’ll save some time at the end, Answer any questions, ask us if it’s really specific for your site and we can’t dive down that rabbit hole.

We’ll get back to you after the show as well. Go check out this recording. It’ll come in the next couple days. And while you’re at it go check out our webinar library. I believe there’s over 100 webinars from Nick UBank on keyword research. Another one that you’ll dig. Also Pam Diner on Sales Enablement.

Andy Cresta, Dina, Kevin Indig on User Intent. When Kevin and I were talking about that before, it was cool. And, there’s so many more for you to check out. Cool. So now getting into our awesome guest. He is the founder of Topic Ranker. He is also founder of Just Reach Out. And one thing that some people don’t know is that you also sold one of your businesses to Google and you also have the PR that converts site or academy where you train people on how to use search engine optimization and PR to their best advantage.

The Meet You Drag Web. Thanks for joining us. Hey, thanks for having me,

Dmitry Dragilev: Jeff. Thanks for having.

Jeff Coyle: Awesome. So what’s now, what’s new? What’s the mission right now? I know that you’ve got a lot of things going on, and I don’t wanna steal your thunder and say them out loud.

Dmitry Dragilev: I so I sold, Just reach out and I’m right now, I started the topic ranker, and that’s pretty much most of my work now is just growing that company.

Topic ranker helps you do keyword research, find problems and weaknesses on Google, and figure out what you should be ranking for. So that’s my world now. We just launched it, so it’s been fun, but about eight months now building it. We have a team. We like seven people now, and just I have to get it out there.

We have some beta testers in there already.

Jeff Coyle: Yeah. At the end of this so the teaser at the end is Dimitri is going to. Some things accessible for anyone watching this recording or this live. And they are limited in their, in the amount of them. So the fact that if you’re listening to this live, you will actually have first dibs on something that I can’t imagine you not doing.

So we’ll get to that. First, what I wanna talk to you about though is, we’re gonna get into the ways that page level analysis or keyword level analysis differs from site level analysis. But I’d love to get your take right now on what are the signals that you’re looking for in a search result or in a competitive landscape that may exhibit things that make it easier or harder for someone to approach that.

And this is agnostic of their existing site. Let’s just generally, what are some of those characteristics that you hone in on?

Dmitry Dragilev: So agnostic of the site. It’s the first thing is there Moz da within the spread of what Moz da ranks on the first page of Google. So if you’re Moz da 10 and you’re trying to rank for a keyword that has a bunch of Moz da 80 or 90, obviously you don’t qualify to be.

But that really stipulates on the topical depth analysis, of course do you even qualify on the topic itself? But beyond that and beyond Moz da, which is the obvious one, we look at reading ease level and Google Page speed insights as well. We’ll look at number of mismatched keywords in the title tag on the search.

So if somebody’s searching for like strategy for keyword research 2022, and everything on the title tag that’s ranking on the first SERP is, doesn’t include two or three type. Those words, that’s a weakness, right? If your content is old more than six months, that’s a weakness. If your content doesn’t have at least a thousand words or two thou a thousand to 2000 words in, in the article.

That’s a weakness. If your content loads more than three seconds, it takes for to load, that’s a weakness. And so we calculate these different weaknesses, like a forum site ranking on the syrup. Like these little things. We catch them and it’s it’s just search results analysis with a crawler that we built.

But it tests things and it just tells you, Hey, you should rank for this keyword, SMS text messaging because A, you’re an authority in it, but B hey, you should, there’s weaknesses and problems in on the search result for you, like you should attack these email, create a page that’s longer, that’s more recent, that loads faster, that has a better reading easiness level.

Jeff Coyle: So it’s taking, it’s taken the a lot of things from experience, and by the way, everything you said gives you at least a collection, a composite collection of signals that one should look for. That’s a list everybody should, take to the bank tries to do those things in a semi, semi-automated or automated way.

But you think about that. One thing I’ll key in on, and I really love it, and I think there’s room to expand it is the signals one gets when you do have a title query to title mismatch or intent mismatch. So how do you think about that now? That’s something that I think about a lot.

It can be a signal of intent fracture. Google doesn’t know which intent should service this. It can be a signal that there’s a larger buyer journey required to be covered. How do you hone in on that with that, just that one characteristic of query similarity to to title sequences. We,

Dmitry Dragilev: so we, we have some AI that tries to extract meaning from the word that they put in, into the search.

And then from then on our version, one of the algorithm was literally just counting keywords. So anybody puts in like SMS text messaging, marketing examples, 2022, and any combination of those keywords should be in the title tag of the search result. If we cannot find all of those keywords, we used to flag it and be like, This is bad.

It’s a weakness. We realized that it’s not so simple because it’s just what if 2022 is missing from there, or what if SMS is missing? From there, it’s still text messaging, marketing examples that people are looking. And so we started like writing logic into extracting meaning from the actual term.

And then grading how the meaning, meaning fits with the meanings of the other other search results that are on the first page of Google. And if the meaning is 80 or 90% there, then we won’t count that as a weakness. That extraction of the meaning is not an exact sign. So we’ve been trying to perfect it in this second version that is out now that we have beta users using.

I have about 15 people in there now that are putting stuff in and are testing it. What I’m finding is people are just like you and I talked about earlier, like they wanna rank because of their ego, and that’s such a widespread problem. It’s just yeah, you’re an expert, but Your site isn’t an expert in this thing.

Like you haven’t talked about it on your site in ages. Yeah, you have a business that’s 30 years old. Yes, you have a 5 million business, but like on the site, you’re not an expert. So just because you put it into our engine, it doesn’t mean that you can actually rank for it. And so we’re trying to let people know that.

But a lot of customers are like, Hey or

Jeff Coyle: something, yeah. Tracking back to two points that I think are very actionable and in unpacking that is, there’s been a kind of a collection, an ensemble approach to this. By hand the people have done for intent mismatch assessment and you’re going from, word checking, word co-occurrence to now trying to look at the.

Meaning of the word and trying to get to that next tier.

We’ve been doing this for, seven years in a different way, where we’re looking at topic models. You also have people who are looking at this from a se similarity perspective. So looking at, putting this word with some other words that are related and how do the search results differ?

But I think we’re all trying to get to the same outcome. And I think that the, there is an answer, and it’s an ensemble approach to these three or four different things with the goal of saying does, do, is this explicit intent and does the query match the pages being returned? So effectively we’re getting to the point where, and I’d love your take on this, is we’re trying to say whether Google’s doing a good job, right?

Yeah. You’re explicit. If it’s an

Dmitry Dragilev: explicit query, yeah. We’re trying to assign the grade to the search result on our side. Based on different factors. So this is just one of ’em, the mismatch of the title, but it’s a big one. And so it, it’s a, it has a bigger weight towards our grade, like internally for this, for the search result, trying to basically automate the whole kind of grading process.

If I think of those graders that, like the Google hires, that like grade search results, right? Trying to automate that process will make it a little better.

Jeff Coyle: But it something I’ve been working on for years and it’s, getting internally, prioritizing that process.

It’s something that we got to a particular point, it produces great results. Let’s go back in and do more research on it. You don’t really know what the there is gonna be, right? You don’t know how much value is this. Level of innovation or research, what that next algorithm that you do, right? That, know, you did the thing.

So that’s something that I’m excited about both of us working on at the same time. It’s kinda cool. Yeah, it’s really cool. ,

Dmitry Dragilev: I like us doing this together cuz I think we can help each other cuz you have the topical depth. And I’m doing like the SE analysis and so if you marry those two, then it’s like somebody can just put in a url, you can tell them, Hey, you’re a topical depth expert in blah, blah, blah.

And then feed those into searching SES for weaknesses. And then literally be like, Hey here are the things that you should write about cuz you’re a topical, expert.

Jeff Coyle: Checking back to part two of what you said that I really hone in on was you, I like to joke around, say that you just haven’t earned it yet, baby.

And it’s, you don’t have, you don’t have it right. You, there’s no chance you wanna own. The iPhone 13 review. Guess what? It’s gonna take you a long time and a whole lot of money. Even if you write the 20,000 word opus with beautiful images that scores a hundred in your favorite 1999 TF IDF tool you, you’re not gonna do it.

It’s just not going right. So you discussed it we think about it as breadth and depth and quality and comprehensiveness and authority. Off page factors, links, social.

What other things do you talk to people about and about how they can more personalize this experience so it’s not biased by what they think they should rank for versus what they actually can, how do you give them that?

Dmitry Dragilev: So we actually, in the tool right now we have a little like disclaimer like, Hey, when you put your topic in, how you a true expert in it? From your website’s standpoint, and there’s a little tutorial that tells them, Hey, what have you been writing about recently in the last six months? How many articles have you covered on this topic?

So if you’re putting in SMS templates for marketing, how many articles have you written on this topic? Straight up? Just tell me and be very specific, like three tens, one. And so if they put in like they’ve only written once about it in the six months, or Hey, this is probably not a good topic for you to put in there.

So I’m trying to get to a point where somebody’s covered this topic on social and on their website at least like 10 times in the last six months. But hopefully way more like I’d love for them to be like, Oh, I’ve done this five times weekly, I do this weekly. I do a weekly post on LinkedIn and I do a weekly blah whatever.

Like some kind of thing about it. And maybe an article every two weeks on this topic. Then it’s Oh, okay, then you deserve to be a like expert in this because you’ve been talking about it all this time. And it’s very hard to come up with unique stuff to say, unless you are an expert in, unless you’re an expert, you,

Jeff Coyle: You’ve angled something that’s really important.

And it’s the nature of what we’re doing here is to say, if I truly was an expert on this topic, what are the concepts that I would cover on this page? But then also, what are the other things that I likely would’ve written about? And so you’re getting to that from a direct question, saying, how frequently do you write about SMS texting for marketers?

Because if they say weekly, monthly, if they say weekly, and you’re like, Okay, most likely they’re an expert. But now you, because there’s only so much you can say. And if you were just a keyword person, maybe you were trained by an affiliate. Hawk, you might only build the one page and move on.

Because anything, you’re only allowed to write one page for every word Dimitri? Yeah. And that, so that’s the difference of what we’re talking about. I love the way you described it cuz you’re actually getting to the answer really quickly and saying, how frequently do you cover this?

And if they’re like, I’ve only written one article about it, You’re like, so I,

Dmitry Dragilev: I actually, before launching topic rank, I wrote a little crawler and they would put in their blog and they would crawl the blog and it will try and find that term on the blog, like through all their blog posts. And it was doing okay, but it was just like the user interface went like this.

They put in MarketMuse and gestation period of a rhino. And like it’d be. You covered zero zero times, you do not deserve to rank for this. And it was like a little bit in their face and I got some pissed off beta users about it. I got four people that were like, I was really like caught off guard.

I really didn’t like the language, so I was like trying to dress it up and I’m like, ah, I don’t know. I’m just gonna remove that and ask them to put in a term that they’re an expert in. But I do have that in the back pocket, so like maybe we can incorporate in there and be like, Hey, you don’t deserve Tori for this cuz you haven’t written in you about it on your blog.

Some people are like, Oh, I have another site. And I’m like, okay, put that site in. There all these edge cases. So I didn’t, so I took that functionality out of there. So if you’re gonna get in there after this call or whatever, that’s fine. You’re not gonna see that. ,

Jeff Coyle: the first version of MarketMuse had the first version of MarketMuse had that.

As well. Really we had the, we had very similar, yeah, the site audits function, which then evolved into our inventory offering. Where now cuz we’re providing it as inform of data, it’s a lot less, you’re telling them that their babies are ugly in a nicer way effectively. And, but we have a release in Q4 that I hope, knock on everything, lock on the wood linoleum.

That has an interesting take on on demand site level quality auditing. And it’s a nice pairing. So it’s not exactly what you’re saying, but it’s trying to get to that similar end. Cuz what you’re saying is you wanna figure out that a search result is potentially an easier for anyone approach and then tune that to be slightly personalized.

Is that, And that’s the best that you’re shooting for here and that’s great. I think that’s,

we’re all searching for the personalized recommendations or. Just a good point of reference that isn’t served by difficulty in the market. And so like how do you feel about the market at large for generic difficulty scores?

Dmitry Dragilev: I think no, everybody overlooks this part that we just discussed. Like people, they just, nobody cares. Do I deserve to rank for it? Yeah, I’m gonna rank for it cuz I got money and I’m gonna, I’m the best. That’s the attitude. And they go into this and they’re like, Not really, but Right.

And so they just force it. They try and force it with whatever, like teams of link builders, they’ll get to the number one, like they’ll muscle it, they’ll muscle their way up there and they’ll do whatever it takes cuz it’s the, But I’m like looking at it and thinking it’s not sustainable this way. Like you need to actually build your topical authority.

And overall the market that I have these 15 people testing. I get emails every day from them right now, and most of ’em are like, Hey, I wanna rank for this. And I’m like, You don’t deserve to rank for this. Why don’t I, because you just started a blog. What do I do? I have a company I need to do the And even these are like SEO agencies that, do link building or technical audits or whatever, and they’ll, they need to rank themselves for stuff, but they don’t have the authority in it yet.

So I’m like, here are the topics. You should start by going after these long term, keywords, low volume, and just writing, like becoming an expert in your unique niche. And there’s

Jeff Coyle: a, I mean there’s a brand. Yeah, there’s a brand of SEOs and agencies who promote that muscle. And get to the top. , even though that is the house built on cards, right? It, and it gets you into very risky situations. And it may work in isolation, or it may even work for a long time if you’ve built this infrastructure well. But there is an end of life on these processes. Whereas building authority through expertise exhibition and telling the story that you really understand things that is, that, long term investment.

And for brands, that’s why I, I joke around about affiliate sites a lot. I love them, right? We all love affiliate sites, but it’s if super microscopes.com breaks, you know what you go do. You go by best microscopes for mom.com and if that breaks, cuz you were too aggressive.

You go buy microscopes under a hundred dollars from mom, right? And you can always pack it up and go build another one. Sometimes you invested more or less, but that’s why those recommendations are so prevalent and throughout get aggressive, go shoot for a really risky topic, but you’re not, they’re not writing the SEO for, for BMW or for, ibm, who they don’t really have the the ability to go, blast through a a high aggressive guess of zone.

But what I wanna, the reason why I say it that way is cause I wanna dovetail this against some of the more recent Google updates. And how, what you’re saying and what I just, that, that soliloquy, anecdote, whatever you wanna call it how that. And I’d love to see, hear your perspective on that. What we just saw happen with PR U or product review update 1, 2, 2 and a half.

The most recent updates and then the helpful content update algorithm and where you think the puck is going or whatever. Sporting I ran,

Dmitry Dragilev: I ran a case study, so as they were rolling out, as that was rolling out, I decided cuz TopicRanker is a brand new domain and so we have no authority in anything, no blog posts at all.

So I threw up a blog post on there and the keyword I decided to go after was which keyword to target, what best keywords to target, something like that. And which I think it’s, which keywords are best to target, and had 300 search volume 300 a month. And I was like, All right, you had da.

40 and 60 and 70 up there. My DA was nothing like zero. And I thought, can I rank there without any links? Just by creating the best piece of content that is like most helpful and start sharing it and improving it. And I got it to like the first page. Then I got like maybe three links to it and it popped to number one.

But with Moz DA four or five, I was at number one. I still number one now. But what I was testing is like content itself, like how helpful is the content and how, like how much of an intent is answered in the first I don’t know, minute of reading of the content. I really tested I had like people read it and give me their whatever, take on it, what they what they did use from it.

I had like people I hired to just tried to read it through and then hired out to, to focus groups and stuff. So I think like going forward like muscling your way up with links is not going to be the way things are going to be done anymore. Like it’s going to be about how the content, like topical depth and the content itself how like truly comprehensive and helpful I guess it is for people.

And how does it stand out from the rest? I have this trouble, like with my customers now, is that they all kind of wanna rank for. Things that are already there. And I always say, Hey, if you went to the, like Supreme Court and you had to prove your decision, make, prove your case and get a d, get a yes from them.

Like how would you prove your case that you deserve to rank here for this keyword? And I think that’s the mentality that people should have. Now with the new update, it’s how can you prove your case in court and do you have a sure case here? Without any kind of paying anybody off. You’re not gonna hire any like weird lawyers to try and pay the judge off.

You’re not doing any of that anymore. We’re completely done with it now. So it’s not, it’s, you really should have a clear case in it. And that’s why analyzing the results, finding the true weakness or problem in the result, be it like the. S the title mismatch. Maybe it’s like the content length, maybe topical depth.

I found like real problems with the search result. Like I didn’t like the content that was there. It was old, outdated. The keyword strategies to, for keywords to target were too generic. Like HubSpot was ranking number one with their stuff about relevance or whatever. So I was like, Hey, I’m gonna write something that addresses weaknesses and problems and see how well it does and it did well.

And so that’s still like the only ranking article on there, but it starts ranking for like keywords to target targeted keywords. Always like different sub terms of it as well. But it was like an interesting experiment, like fresh domain, no links, throw up the best piece of content, promote it on social, promote it through email and see like how well it does,

Jeff Coyle: and in those scenarios you can have longevity as well in certain blocks there. If you put. And you’re doing it because there, there is something that’s telling the world that you are about that for one reason or another. So I would argue that while it’s a brand new site, there are signals that says if Demetri is behind this site and it’s connected to the web, it’s highly gonna be related to on it too, some PR and SEO value.

So you’re getting some co some bias, but it’s not, it’s still a good experiment because it’s a new site and I’ve seen this happen time and time again. So I work with the travel site who put out another entity in another location and there is some glow off of it if it’s done well. So I think that’s a great example.

The and I’ve done it, I’ve done it with from scratch sites as well. It’s a good way of testing. Waters on various strata. So this is, take this to the bank on your strategy for brand new sites. You wanna build predictive strata at various levels, and you can publish individual high quality stuff at each level and see how it’s moving as long as they’re cohorted.

I know I’m getting a little bit esoteric, but you might start to then become predictive with a brand new site in that way. Oh gosh, I can’t believe I said that on a webinar. But yes, that’s a good way to launch a brand new site. And so anyway, moving on.

So the helpful content update is a an interesting one also because it is a machine learning classifier of individual pages.

Using a composite score of all these things, right? So it’s using the user behavior, it’s using characteristics, it’s gleaned from the site, and it’s classifying an individual page to say This is helpful or it’s not. And then it’s collecting that amongst a site and saying, You have X amount of your content.

Absolute, or X percentage of your content is low. It is not very helpful, and that can act as a weight on your entire site. So the reason why I bring that up is, it’s basically saying something you said before is to say one bad page. An entire section of bad pages or an entire section of pages that aren’t relevant to what you do right?

Can actually drag down all the good stuff. It’s the first time Google. So I love your take on that. Yeah. Brian

Dmitry Dragilev: Dean got this, right? He did every page is useful. Highly valuable and well maintained. So like websites with high amounts of unhelpful con like websites with a bunch of articles, like all those SaaS blogs with hundreds of useless pages.

I just go delete that shit. Like it’s, we’re gonna get into that on your side should be very useful. That’s my

Jeff Coyle: Get back to the whole thing. You know I’m not leaving that alone. So two pieces of that. So every page being high quality and useful and having a purpose, nobody’s gonna debate that, that’s obviously, But the, is that saying that you should be looking at one page, one word. as a target, or you should be looking at pages that rank for 7,000 words and not acting on them, which is where I’d say the person you referenced in there gets it wrong. Cause you, you’re so susceptible to risk.

Somebody could publish content on a better intent. People like you who are analyzing the SERP and chop down the tree or chop down floor by floor of the skyscraper, if you wanna use that language,

Dmitry Dragilev: his skyscraper, as long as it’s in each website overall, should have a clear and primary focus, right? No longer are doing like SEO and marketing and affiliate and all these other things.

Like you need to have your niche and that needs to be your topical depth and that’s what you’re really good at. So if you’re going to go after like your blog clean up, Then I would say marry that with your primary focus. And then see does this page fit with your main primary focus? And if it doesn’t, maybe it’s time to part ways with it and let that go, because it just doesn’t, it doesn’t really serve you anymore.

It used to be in where you can have a blog about all sorts of things. But now I feel like with this update really should be one main focus,

Jeff Coyle: yeah. If you’re gonna go after it, it’s gotta be a lot of heft. You can’t, one day, it can’t have a small blog that is about kitty cats today and tomorrow it’s about, nuclear vision.

It’s not likely to have much of, a much of a move. But I wanna go back to the the pruning connection. So deleting what you don’t think is on target for topically versus deleting low quality. versus redirection versus repurposing? What’s your take on that? You made a pretty blunt statement about deletion and I’m very passionate about deletion.

Yeah. Usually I’m on the other side of the of the table. Cause I believe, no matter how ugly, I can make it pretty but what is your, what’s your take on on deletion as a process or a technique? Yeah, so I

Dmitry Dragilev: go through my like my focus and then my thing is seo, keyword research and pr.

And so like on the JustReachOut blog, I’m still involved with that team. Anything that is not around pr, like we’ve written about all sorts of stuff is a candidate for deletion, unless they can be changed to be about PR easily or old content that is crappy. If it’ll require us, let’s say two grand or one grand to really change it into some type of really great piece of content, then maybe we’ll and salvage it.

Maybe we’ll keep it. But honestly, most of the time we’ve been just this is not performing. It is not really aligned with our topical depth or like our content focus. Probably we should just take this out. Can we align it with something maybe? So we have a list of targets and we’re like, Okay, does this marry towards that?

Great. Then we can go and go through the process of changing that. But we it’s old content which has accumulated decent links and we can marry it with a new intent or like a keyword that can actually, we have a weakness for great. But if If we just can’t marry it to a specific keyword that has a problem or weakness on search results we’ll probably let it go.

And if there’s a lot of links to it, we might try and salvage it. We’ll try hard to try and marry it into a content bucket that, makes sense. But we we have one, one article about like website visitors and tracking their IP address and then trying to email them and sell your stuff to them because based on IP you can get their URL and all that.

And that topic is not really related to PR is not really related to seo, but it’s got seven or eight decent domains, like really nice domains pointing to it. And so what we’ve tried to do is marry that into like, How to find leads from seo, that kind of thing, just to make sure that it’s in the SEO space.

And we kept it because of the links. But usually it’s like the first pass is let’s delete all this stuff. Then it’s like the second pass is Wait, let’s keep like this stuff and delete that. And then the third pass is 75% of this stuff. I guess we’ll keep, because we have links for it, let’s just work on redoing it.

But nobody’s really excited about it. So it’s on the back burner all the time. ,

Jeff Coyle: Okay. No, it is just shows there are so many different processes for evaluating. The one thing I’ll touch on though is you are putting a lot of experience led decision making into that process, right?

You’ve done it a lot. You know when what I struggle with is people, when they sort descend based on links and they sort descend based on traffic and they delete the bottom. Because it’s like pulling teeth outta your jaw. There are pages that are really important to your topic authority that don’t get any traffic.

And so if it fits, you gotta you’re making four or five statements in one. You’re saying, is it a fit for this topic? Does it tell the story of expertise? Is it high quality and does it have all these traffic signals? You’re not deleting something. We’re adding it solely based on the fact that as links or not, that’s a line of designation. I think people with a lot less experience in this space often will hear that as delete all the stuff with no traffic and it causes disasters. I, trust me, I’ve been in a lot,

I’ve got a really good question. Two, two really. Oh my gosh. So many really good questions.

But one of them is the, I’m gonna, I’m gonna say it, I’m gonna read it the way that Delinda wrote it. And then I’m going to rephrase it potentially to be two, two possible answers. Blogs about events that happened with your team are not good to write. My question, I’m gonna directly answer that one if you have it.

But I’m gonna rephrase it a little bit and say, So where does temporal content fit into a strategy? Do you always have to morph, temporal? And when I say temporal, like moment in time act things about what happened yesterday, Something that’s newsworthy temporal object that could go out of style.

Do you always have to morph those into evergreen or do they have a place living as temporal and then answering the other question like, could I and more specifically blogs about my team, is there a place for that? I think it all just

Dmitry Dragilev: builds topical depth, so it’s good.

Jeff Coyle: Yeah, exactly right. , It’s the opposite.

Yeah, it’s good. It’s good. It tells the story that you actually have a business and you work in the space and the temporal stuff tells like you’re a real human that writes, You’re not only writing, what is X alternatives, X versus Y the other thing is you’re gonna watch for it. The, I think to your question is if the site that you’re competing with I’ve been trying to systematize this.

Maybe Demetri can automate this. If the site that you’re working against on the topic only writes pages that are targeted, sort, descend for search volume on the keyword. Oh man, those are my favorite. Those are my favorite to beat. But yeah, you should only, should not only be publishing stuff that’s In that one vein cuz it’s gonna cause you problems down the road.

What do you consider some signals of being, and this is also from delinda of being a good topic. I have really hard time writing many blogs on the same subject. I’ve got a good answer for this. But what do you think? So writing about, you have one topic, how do you write a lot of different blogs about that?

Dmitry Dragilev: I would pick specific queries, like very specific how to whatever like you think is the thing that people need to learn about it and say someone reading that content leave feeling like they have the most satisfying experience. Have they learned enough about the topic to achieve their goals?

, and you should say no when you look at all the results on first page of Google, when you read your own, you should say yes according, like generally speaking, but you should also get other people’s opinion about that. When you’re picking, and of course, like you need to be passionate about this topic.

But don’t, I wouldn’t repeat yourself about the same thing over and over. I would find problems and issues about your topic over and

Jeff Coyle: over. Yeah, I think that’s a great piece of advice. It’s getting into I, I like to think about it. So you’ve got three dimensions here. Three more dimensions to think about.

One’s the audience, who’s it for? Do you only have one audience or do you have many? Where in the journey is the person that you’re writing for? Are they a beginner? Are they experts? Are they in early stage awareness, consideration purchase, troubleshooting. Are they customers of your product and you’re writing for them?

And then you can matrix all that into industries or Something that only an expert would know about this and how it applies to this. So I just gave you five or six angular thoughts. Now you can mix and match them. And that’s a good way to be thinking about it. So you don’t get, Oh, I’ve written this definition.

Do I just write, I write another definition. So yeah, those are some ways to think about it. So what do you think about kind of Dimitri for buyer journey content? How do you think about that? I think it tells the story of being helpful, right? And for brand

Dmitry Dragilev: notification, I have to go plug in. I think battery is 9%.

Jeff Coyle: 9%. Okay. Yeah. You better plug in.

I’ll ask and answer the next question why you go plug I’ll go get it. Richard what’s your opinion on content length? I find that some pages optimized for length feel wordy and you’re not as helpful as shorter focused content. Does the helpful content update factor in wordiness as a negative signal?

Ah, that’s a great question. One, one really great signal of expertise is conciseness how concise something is. So if a long page is about a lot of different unrelated concepts or unrelated intents, but it meanders across hitting on a bunch of keywords, that’s typically a signal of less expertise.

And so that’s where you gotta really watch out for. If you’re using correlation SEO theories copying your how to copy your competitor or you’re copying your competitor you can’t really answer the question about the h the helpful content if it uses that as a signal, cuz that wouldn’t be published.

However there is a list of guidelines on the the Google site that walks through some of the factors that could be involved in this assessment of a page. And just something to think about. It’s definitely a composite, an ensemble approach. It’s machine learning approach to evaluation. So what does that mean?

In plain languages, there’s a lot of hypothesis hypotheses of things that could equal a negative experience. And they get validated or not validated. So if you’ve got, they might throw in, I always use this example cuz it was a real one when the panda update came out in 2011 was someone could sit in the room and say, I think that all red sites are bad, right?

And then they would process that hypothesis against all the data that they have and all these factors. It would shoot at the bottom and go, Actually no, we can’t validate this. But what you’re thinking about there’s a lot of, so somebody there could program, hey, If this page is on lot about, a lot about things and it’s too wordy or it has these specific characteristics, it could have the same outcomes as something that they’ve already validated as bad.

So a good, sniff test there is it providing the value that you need? What are the signals that might happen? If somehow this becomes a way to detect over time, they’re gonna move the needle on the, looking for that, right? I wouldn’t say in this case that it’s something that’s probably going to weigh down your site specifically.

But if you have a collection of sites, pages on your site that are all like ultra long, don’t really exhibit expertise, it’s gonna be a problem over time. It may not be today, it may not be tomorrow. But. You’re not actually helping people with that content. And there’s probably a better strategy too, you probably could get to, could build that into a more appropriate cluster of slightly smaller stuff, answer more great questions and have it be more actionable in practice.

That’s gonna work better anyway. And those are some things to think about. You shouldn’t be writing long pages for long pages sake, or because someone told you that something like outlandishly Large gets you there. I use an example from a real practice. There are news organizations out there. One of them is The Guardian I’ll actually name them, who with a fast moving news event will typically have a 10,000 plus word ongoing page that they just augment other news products that are out there.

See that and they use that to their disadvantage and they publish aggressively into clusters temporally to basically chop down those trees. And there’s other examples of this in practice too. So if you’re publishing long, just for the purposes of publishing long, you’re gonna create a lot of problems.

Will it influence you on the Google Helpful content upgrade update? Nobody could answer that question unless you work within the halls. So Dimitri, long content for long content’s sake, I talked long enough to, for you to get your plug and work. Yeah, no, thanks. Yeah, ,

Dmitry Dragilev: I would, you know how I would think about it is Healthy Hearing is a website, which you should check out.

They were trying to rank for hearing aids forever. , it’s a term that they have 200,000 searches a month, and the. Creating the most comprehensive guide around hearing aids is like crazy impossible. Like how long is that gonna take you? Like an eternity, right? And so what they started doing is going after batteries and hearing aids in water, hearing aid design, like all the subtopics that people are searching, try to become the authority on that one topic with those smaller sort of like hits.

They might be like 10,000, 5,000 a month, 518,000, 22,000. And so as they grew and as their traffic grew with those subtopics of that cluster, then they work their way to becoming the authority for the main term hearing aids. I think that’s a healthy way to work towards something like that. And not trying to just, let’s do a 20,000 page word article on hearing aids or whatever your term is.

But let’s become a topical authority gradually around this topic and let’s hit those smaller keywords one at a time and keep building our authority in the space. And you know what to do if you’re hearing aid batteries lost or what, whatever It was just, And so they have the most, if you check out the site, it’s just a help section, but it’s like a beautifully designed help section with just all these like technology related like articles around hearing aids.

And eventually they do rank, if you put in hearing aids, I think they’re in the top, the five or something for hearing aids now. But but it’s just I like that example cuz it just communicates like how I think people should be

Jeff Coyle: doing this, you’re thinking empathetically, right?

And I, and the great answer to, and Richard, thank you for the question. Great answer to this is you have a lot of different learners in your audience. There are people who love, who will read if it’s great and it’s 20,000 words Oh my gosh, this is the best resource ever. That could also become an ebook.

It could also become, a honeypot, it could be a number of other things. Those are other ways people might consume it. It could also be a collection of audio recordings or videos, so it also could be a cluster of content. Maybe you take your long form content item that’s generalist and you build your cluster for people that maybe are a little bit more advanced or a little bit further down in the funnel, right?

So always be thinking if you’re gonna do something long, make sure it’s not long and terrible. That’s one. And if you actually read it, and you could make it through and it’s, you feel good about yourself after having gone through it, or you weren’t just Oh gosh, this is terrible. There’s some pony there.

There’s a pony there. In that, having that and having other versions and repurposing practices. Now the comparable I’ll give you is transcription. So the maturity model for transcription is of let’s say a podcast or webinar. I’ve got a summary. I’ve got show notes, then I have transcript.

The transcript is for a two hour podcast. Could be, tens of thousands of words, right? You think, Oh, should I put that on the page? That’s the same logic path. Yes, putting that on the page is valuable, but could you morph that into a lot of additional value? How would you do that? And can you an annotate, annotate that?

I think that’s the answer to your question, Richard, that you didn’t ask. All right, cool. Thank you so much. And Delinda you are welcome. I got another good one. And then we’ll get into kind of wrapping up our thoughts, but keep asking them how can I get a replay? Next two days you’ll get an email.

Megan awesome. I love the site that you work for. I know it very well. We’re launching a new narrative for our business and the keywords related to the narrative have low search volume. Oh, this is a great question for us, Dimitri. Our previous keyword strategy had high search volume keywords with more commonly discussed topics amongst our prospects.

We wanna integrate this with more popular words from our previous strategy. What percentage of our blogs should be catered to our new keyword with more popularity versus this new frontier with low volume? Oh man, what a great question. Thanks Megan. Okay, Deri, what’s your take? Old site. Old site, high demand research based on demand.

They’re redoing the site. They’re not throwing anything away naturally. The new site’s going for some topics that don’t have as much demand. What do you what’s your take? How do you think about that? I

Dmitry Dragilev: think the topics that you were going for before and the topics you are going for now, topically should be related.

Like your topical depth profile should be, re related. If we’re going and we are doing like a 180 or 360, like this is not a good thing, we should, but if it’s in the same realm, you were just pivoting and saying, All right, we’re going, We’re not targeting SAS anymore. We’re targeting startups, or we’re targeting different kind of startups.

We still in the startup space and still startup topics. Then we’re in a good spot. So that’s like the first thing to just look at. It’s are we going to be building the same type of topical depth on the site? And what kind of topical depth does this site already have? How is it aligned with it?

That’s like the first thing I would look into before making these changes. And then going after low search volume versus higher search volume. Like of course, like everybody wants the higher search volume. I would look for weaknesses and problems. With that specific term. I wouldn’t really qualify it based on how many searches it gets.

Sure, that’s a good thing to note is a factor. But for you to rank Google doesn’t. Care too much. Like it cares. What really matters in this whole thing is there a problem or a weakness for the search result? And are you topically aligned to, do you have the topical depth authority to rank for it?

And if there’s a weakness or a problem with search result, like the ones we discussed, and you have the topical depth in that space and the expertise, then you probably qualify for it. So I would think about it that way and then use the metric for the search volume as to make the decision, of course, which one you should target as a factor there, but just make your decision based on those things like topical depth and weakness on the search result.

Jeff Coyle: Yeah, I think that’s really reasonable advice. I have a blog post, which I’ll link to in the email that we shoot out here called Keyword search Volume. The Search Volume Illusion. Which you should definitely should read Megan as well. It’s about how people using search volume only as their North star can get them into a big bag of problems.

But great advice from a great advice from a tactical perspective, how semantically related is the new topic to the current one. So do your six degrees of Kevin Bacon research on this topic. Look at the relatedness, semantic relatedness. There may be some bridge topics you need to write about.

So you can’t just write about kitty cats and the beach and then write what cats you can bring to the beach and you’re not all of a sudden about the beach, right? You can’t go from cats to the beach, right? Just by bringing cats to the beach is the example I’ll use. Sometimes I use beers at the beach just cuz you know a lot about beer and a lot about beach.

You can’t bring beers to the beach. You can write about that, but it doesn’t mean you’re about both. But if they’re semantically similar, you can get away with a bridge. So you may need to write middle bridge content as part of this new site strategy to help you not die. If they’re not semantically similar.

If they’re semantically similar, you are gonna get away with it. So those are just things to take. I would not shy away from something because a search volume, if it fits with I would treat it like a new product launch though not a rebrand. And you’re gonna have more success. So I have this happen all the time.

B2B technology company owns this one topic. They bought this other company and they’re like, We wanna own the topics that company has, right? So just treat it like a new product launch and you’ll have more pragmatic success. That’s a great question. And I’m gonna give one more and then we’re gonna talk about your promo for the new solution.

But Bogden or Bogdon, not sure about the pronunciation. Many apologies. If you start a new blog, is it recommended to start with traffic strategy and target the keywords that have the potential to attract topics? And then. Switch to pillars and clusters or start directly with pillars and clusters. love this question.

I wanna know your take on it. Do you shoot, do you, I’m, do you long tail into the pillars and clusters? Or do you start with the pillars and clusters

Dmitry Dragilev: then? Long tail pillars and clusters? I wouldn’t start with potential to attract a lot of traffic. If you’re starting a new blog, like sure, you want a lot of traffic.

Everybody wants a lot of traffic, but it’s a new blog. So if you go after a keyword that’s I don’t know, 50,000 searches a month or a hundred searches, like that’s the north star illusion on the key on the keyword volume. That blog post would help you. It just, you do not want to be like, diss way that like just pick a keyword that has a lot of volume and traffic and think that you’re gonna get it.

It’s a new blog. So with a new blog, it’s important to build topical depth authority. You Google doesn’t know anything about you book then, like it doesn’t know what your blog is about. And if you put in one article or two articles trying to go after the North Star of a hundred thousand volume search term it’s not going to give Google a very good warm and fuzzy on who you are and what you do.

Most of the time you might not rank for it for a long time. So with a new blog, I think content and clusters pillars and clusters would be the best kind of approach to start. And you can start gaining traffic one at a time. And it doesn’t mean that you’re only going after 50 searches a month.

You might be going after a thousand searches a month, but you are still building out that topical depth authority on that specific main topic like hearing aids. And underneath you’re still building out like, batteries for hearing aids or whatever it is for hearing aids that you’re writing about, but you’re inching on to, like trying to rank for something that guess a thousand searches a month, you’re gonna get whatever, like 10 clicks a week or whatever you’re gonna get.

You’re not gonna get crazy traffic from it. It’s just a long game, like nobody’s doing this kind of right away. And so I see a lot of customers come to us like, All right, I want traffic. How can I do this the fastest? This no wrong way to think about it. Like I, how long is it gonna take me? I’m like, wrong way to think about it,

But really I’m,

Jeff Coyle: I’m gonna cut. You. Set expectations appropriately, right? How long it’s gonna take. A good way to do that is a process called competitive cohort profiling. There’s a post on the MarketMuse blog called Content Strategy Example, Combining competitive analysis that walks through that process and go check that out.

It’ll allow you to set expectations for how big those pillars and clusters have to be. So then you can decide whether you want a long tail on top of it. So I love Dimitri. Dimitri basically said wrong question, but it’s right question cuz it’s the wrong way to look at the right question cuz you have to set expectations and you have to budget, right?

It’s just so hard when it’s a brand new site. So there’s some ways that you look at it. And wow, Dimitri, so many just insightful things, great questions from the audience, and they always are. Tell us about the promo with Topic Ranker for the launch. You listening to this can be an early evangelist for topic ranker and get a direct line to Dimitri.

You always have a direct line to me, see how I did that? And you can shoot me a note, Jeff MarketMuse dot com. Book a demo. Do a audit. But what is the promo for topic ranker? And how can they get going with that?

Dmitry Dragilev: We’re we launched our beta this week, and so what I’m doing I have very early access pricing, 59 bucks a month, $129 a.

To get in there and start testing the product getting keyword recommendations. And so for the first 15 people I’m living in 15 people at a time. We already let 15 in, I’m gonna let 15 in more right now, and then we’re gonna close it and we’re gonna do a closed beta for from then on. But you’ll be locked in for the $59 a month, or $129 a month for the lifetime of your account.

So when we increase prices, we’re gonna go to annual only later you’ll still be locked into that. You also get one phone call with me and email support from me choosing keywords, figuring out what your clusters should be and all that stuff, just because we’re right now in the process of testing the product and I need to be hands on involved with our customers to figure out how product is working.

So you get basically one-on-one consulting from me for $129 a month. but I need to do it, the product.

Jeff Coyle: Ok if everybody listening to this live doesn’t just close those beta, I don’t know what happens. But thank you to Mitri. Also, a little quick plug. We’re both gonna be at Roadium next week.

If you’re not familiar with what Roadium is, it’s a mastermind community that we’re both a part of. If you are a founder or an owner go check it out. If you’re interested, reach out to one of us. And we’re both the leading sessions mine about the helpful content update and yours is about

Dmitry Dragilev: keyword research with weaknesses in research

Jeff Coyle: results.

How lovely is that? It’s like we plan this thing. Alright, thank you so much for the little bit of extra time. I’m glad we, your battery didn’t die and thanks for everybody for joining us. Go book a demo and go check out the closed beta for to crank right now. All right, thanks again. See ya .

Tweet
Share
Share