Hi everybody, and welcome to the recording and recap of Page One Power's Link Building Expert Panel Discussion.
Our panelists shared valuable insights from a variety of unique perspectives, and we can't thank them enough for sharing their time and expertise with us.
Hi everybody, and welcome to the recording and recap of Page One Power's Link Building Expert Panel Discussion.
First off, thank you so much to our panel of industry experts which featured:
- Casie Gillette, Director of Online Marketing at KoMarketing
- Pete Campbell, Managing Director of Kaizen Search
- Paul Madden, Co-Founder of Kerboo (formerly LinkRisk)
- and Nicholas Chimonas, Director of R&D at Page One Power
Our panelists shared valuable insights from a variety of unique perspectives, and we can't thank them enough for sharing their time and expertise with us.
In this webinar, our panel discussed multiple topics ranging from leveraging linkable assets, link analysis, the upcoming Penguin algorithm update, and more. There were a number of key takeaways and important points, so without further ado let's dive into the discussion.
The webinar ran approximately 60 minutes during which the panelists discussed seven questions.
The first three questions were decided upon before the webinar, to act as launching points for the discussion. All other questions came in real-time from our live audience, and were chosen for their relevance to the conversation or their merit.
The following recap is a summary of the questions and answers discussed during the webinar. These are not direct quotes, but rather paraphrased responses. If you would like to hear specifically what was said a video queued to each question will be provided in the corresponding section.
I encourage you to listen to the audio for questions you are particularly interested in - the panelists discussed each question in great detail, and my summation of their answers will not capture everything said.
- How do you identify and leverage linkable assets?
- How do you analyze links on both an individual and domain basis?
- What are your high-level opinions of the upcoming Penguin algorithm?
- What is your best tip for creating linkable content?
- If you're not doing well in organic search (due to potential negative SEO) and PPC is extremely competitive/expensive, what would you advise?
- What are you opinions regarding the future of manual link building?
- What are the best tools for checking backlinks?
Let's see how the panelists answered these questions.
Question One: How do you identify and leverage linkable assets?
The discussion of question one begins at 2:27.
The question lasts until 14:43
Nicholas: Identifying and leveraging linkable assets is a big problem in the link building industry, because you often have clients with misconceptions that you can build links directly to their product pages. However, nobody wants to link to these pages unless you're giving them money.
One way we circumvent this issue at Page One Power is by building out content assets on a client's domain that we can build links to, and then we pass link equity through the site with internal link architecture. However, even finding what assets exist on bigger sites can be difficult.
Casie: Some clients have a number of assets, and some have nothing. I like to start with what they are already sharing and what is already driving links and shares using tools like BuzzSumo and Open Site Explorer. This process can point you in the right direction without forcing you to dig through the client's entire site.
But in the case where they don't necessarily have anything, I like to look at their competitors. Of course, you don't want to blindly copy what the competition is doing. But if competitors have already done the legwork in figuring out what customers want, it's a really good place to start.
Those are my favorite two starting points when thinking about what can we use to drive links to the site.
Nicholas: Often times you have something that has performed well and naturally earned some links, but even then there are opportunities left on the table because links don't build themselves.
I'd also be interested to hear what Pete and Paul have to say.
Pete: For me, it's not too disimilar. I'm always looking for diamonds in the rough - either by using BuzzSumo (shares), Majestic (links), or Google Analytics (high time on site). You can look subjectively, or scientifically at what content looks ugly, or not in-depth enough and can be expanded or improved into a more rich resource.
An example I can give you is something my colleague did, where he turned a long, boring blog article into an interactive tool. You can take what most clients have, such as a long useful resource that is text-heavy, and give it a nice design and think about how you can make it more in-depth. My process for that is using the aforementioned tools to find these resources.
Conversely, if they don't have these resources I look to competitors to produce new things. For example I use BuzzSumo to research the top three or four competitors to try to understand what topics and content formats work best. You can use this information to brainstorm content ideas by knowing what works well versus what fails.
In our experience, success comes down to the execution of the content itself. Whether it is in-depth enough or well-designed - people don't want to read 500 words, they want something light, succinct, and to the point.
Nicholas: Really good stuff Pete, thanks for sharing. What that made me think of is Brian Dean and his skyscraper technique. But people tend to miss that they can apply this technique to their own site. You might already have a great whitepaper or great research on your site that's just not put together right for the purposes of link building - you have the data, research, and content, but it does have to be designed a certain way to be a linkable asset.
Paul could you tail us out on this and give us your knowledge?
Paul: From our perspective, what we've been doing recently is looking at tools like BuzzSumo and trying to derive some algorithm that understands influence and reach.
We want to be able to direct people to the piece of content that actually has some influence and impact, rather than just the most signals. While looking at the content we analyze, we've realized that you really need to look at the shares and links and consider how many of those are genuine and natural, and how many have been manipulated and seeded into place.
We spent a lot of time trying to understand how to differentiate those pieces of content that get genuine, long-term coverage, and how many have got the boiler plate style seeding type responses that we see all the time. You have to look deeper than just the total number, you have to look into the quality of what's in there as well.
Nicholas: Can you give us an idea of the kinds of flags that your algorithm is using to assess link quality?
Paul: The LinkRisk algorithm takes into account over 150 signals and layers them on top of each other. On the risk side of things we are trying to derive the intent by which the link was originally placed, so if it was placed for manipulation of Google's algorithm that is what our algorithm is designed to detect.
On the influencer side of things, we are looking for three different angles - the reach of the post, the velocity of the post, and the influence of the post. We're trying to derive from a final score that allows us to make sense out of the madness that is link data and share data.
Nicholas: I think this is data that is going to be more and more important for us in the SEO industry.
The past few years as content marketing has taken off, a lot of the SEO industry at large accepted the assumption that if you put great content out there that gets shared a lot it will get linked to, but recent studies from BuzzSumo and Moz have shown that's not the case at all. We need more data on what's really happening with sharing and linking, and I'm really excited for what you're putting together.
Paul: We can back that up by saying sharing and linking depends on the content type. So if the content is very shareable, it will get shared. If the content is a resource that's valuable to attribute in some way, it will get links. It is very difficult to find pieces of content that do both.
Casie: That's such a great point, having something that can be attributed to.
We work with a lot of B2B companies and they put together a lot of reports, case studies, and whitepapers that are quite valuable. But the challenge is they want to use those to generate leads, so a lot of times they're gated. Those are the types of situations where we try to explain those are the things that are really going to drive links because they actually have solid data. Those are the types of content that drive links rather than just shares.
Question Two: How do you analyze links on both an individual and domain basis?
The question begins at 14:58
And the discussion ends at 29:19
Nicholas: Let's talk about what kind of metrics and measurements we use to analyze links at both an individual and domain-wide level.
Paul: The way we an analyze links is by trying to derive intent, and as of a couple weeks ago we have scored roughly 8.2 billion links, so we have a large data set. And from that we try to take as many signals as we can, and layer those signals on top of each other to derive how a link got its placement.
We look at tactics from the past and try to derive why they were done, and systems that have been abused and are problematic in terms of risk. We also look at relevancy, positioning, frequency and all sorts of things. We try to translate that into how the link is viewed by Google, and if it is passing value. If it is passing some value, we want it. If it isn't passing some value, what's the point?
So we take all that we know about quality and intent, and try to derive whether or not Google is passing some value through the link. We all know the vast majority of links probably don't pass any value, and it's only the links that pass value that we're interested in.
In fact, we've seen a lot of cases where clients remove links and their rankings improve as a result. But we do a lot of work around deriving the intent, and everything boils down to intent. You should look at Google's link schemes page because that's the page you're being judged by and anything that could apply to this page is what we look to cut out of the equation.
Nicholas: I agree, it's all about intent. When we build links we think about that too. Almost none of the links we build are completely in our control, they come through an outreach email from somebody on the other side of the web putting it up. And that is how we avoid penalties, by letting people put up links as they would put them up.
Paul: You also have think about how Google views links.
For example, Google seems to hate the forum link even though they are user-generated, genuine, useful, editorially given content. They seem to be pretty bad at determining what is manipulative and what is not. So we end up advising customers to disavow and remove links that otherwise have no reason to remove. You really do have to think about how Google in their ivory tower think about these links.
Pete: Well Paul got me thinking and reminded me of one of the challenges I've faced.
We have examples where we produced content for clients that was very successful in terms of coverage (mentions in national newspapers) and referral traffic, but it didn't move the needle at all for the client's visibility. And it shocked me because I have always been of the opinion that the if I get high authority links the rest will take care of itself. Where every one authority link I get will make up for the twenty paid links that competitors have.
And I still find it interesting today that when you look at certain search results, you can't deny the influence of a link that has keyword-rich anchor text from a mid-to-high authority domain is still far more influential even though that link is for manipulative purposes and was not editorially gained.
But what I was originally going to say was that we use backlink analysis to spot link opportunities. Quite often what we do is analyze three or four competitor backlink profiles to identify links that they share but our clients don't have. It's an old-fashioned tactic, but it's worth doing again and again because it works quite well.
And the other element in terms of setting the stage, we quite often look at the gap in page linking root domains between our client and competitors. More than anything we use this to manage expectations in terms of how competitive the niche is.
Nicholas: You're absolutely right. I find it to be true as well that the keyword-rich anchor text link is still very powerful, still relevant. It works great if you don't have over-optimization already.
Paul: I think that is very true, that the blatant stuff still works and it's about where you draw the line. We now have a line between not moving the needle and Penguin, and we live in the middle of that. But anchor text links still work, god bless anchor text links.
Casie: The way that I look at links, in terms of analyzing them and determining what is a good link is, by approaching it two to three different ways.
There is a whole idea of finding these niche sites that may not be the highest authority links, but they are really relevant to your customers. These links drive relevant traffic and they drive leads.
There is an idea that you need a broad spectrum of links and a ton of links, but that's not necessarily true. For a lot of our clients, we're just trying to get them relevant links that make sense for them.
I also think there's that idea of tiers and getting a client general press coverage might not be as relevant, but is a high-authority. So you have those news sites (Forbes, Inc., Entrepreneur) that are just good to have a link on because they will help you overall.
So I look at links in those two ways: in terms of niche and actually driving customers, and how it is going to help you overall with brand awareness and authority.
I also wanted to touch on the idea of forums, and how Google views forum links. The reason I say that is a lot of our clients in the technology space get a ton of customers from sites like Quora, SpiceWorks, and other tech forums because they have to be there.
We can't always be concerned about if we will get in trouble for putting a link here, and we try to find forums where that won't happen, but at the end of the day it comes back to where are your customers and sometimes you have to put that over the actual SEO value of the link itself.
Nicholas: It's also valuable to diversify in that way just in case something happens to your Google ranking, it's not your entire source of traffic. I always value where the human user is over what Google might be flagging.
But I want to share my one actionable tip for link analysis. It's about internal link building - actually going through your site and finding which pages are relevant to put internal links to help other pages that aren't getting enough equity from links. (Nicholas referenced the process outlined in this post by John-Henry Scherck)
Question Three: What are your high-level opinions of the upcoming Penguin algorithm?
Question three beings at 29:23.
The responses to question three last until 40:17.
Nicholas: I would like to get some high-level opinions on the upcoming Penguin algorithm which is allegedly coming out before the end of the year.
Pete: For me, it is a sigh of relief. I'm kind of glad this is becoming a real-time element because we've had customers come on board who have spammy backlink profiles and we have to remove links and submit a disavow file. And there have been certain instances where we know we're not really going to see the needle move until we a Penguin update rolls out, and it's our best guess as to when that will happen based on past Penguin roll outs.
There have been instances where I've had conversations with clients about rebranding or changing the whole domain name, because the penalty won't pass through to a new domain name. It's ridiculous to have those conversations because we know it (a Penguin update) will potentially save their business. There are certainly businesses we work with where they have no idea about the damage their previous SEO agency was doing to their business, and they just want to resolve it. So I think it's been a long time coming.
In terms of how it will change, I think doing regular backlink analysis will become more prominent. And I would hope that negative SEO will start to die off a bit.
I think it is common sense and I don't know why Penguin wasn't real-time from the start, so it's a big thumbs up from me.
Nicholas: 100% agreement across the board. It can be such a battle with clients to say, "hopefully what we did will work a long time from now if you're still a business". It can be really sad, I've seen clients who've had to lay off staff because they hired a spammy SEO and now they're suffering from Penguin. I hope to see it (real-time Penguin).
Casie: I hate the idea of a disavow file - I hate the fact that we have to tell Google "Take these links away". And one thing that I hope this rolling update leads to is a better sense for Google of which links they shouldn't count, and then they can discount those links.
My friend Aaron Friedman wrote an article that talked about the idea that if you get hit by Penguin you can disavow and clean everything up and then maybe in a year your site will bounce back. But I don't think that's necessarily the case because Aaron brings up this great point that those artificial links were inflating how your site preformed.
So I think with the ongoing updates it gives people a better sense of where their site actually is. So you're not just hoping and doing all these things to find out in a year that your site is only three-quarters or half of where it was. Now you have a faster opportunity to build that back up.
I really hate that we have to tell them here are the links that just don't count. I think it should really be on Google to figure out what not to count versus penalizing someone for links.
Nicholas: That has always been my opinion as well. I find it incredible that Google is out there saying negative SEO could never happen because they understand the intent of a given link. There's no way! Anybody can go out there and buy a bunch of links and make them purposefully look manipulative. If you know what you're doing, if you've seen a lot of manual penalties or Penguin devaluations, you know what causes these things to be triggered.
Negative SEO is real. And I hope Penguin becomes real-time and I hope it quells the negative SEO industry.
Paul: As far as the upcoming Penguin update--or "Great Sales Day" as we call it here at Kerboo--we are very much looking forward to it. I am very keen on them getting it baked into the algorithm because it's unfair that people have to wait.
When I'm asked if I would rather have a manual action or be under the influence of Penguin? I'd rather have a manual action please, because I can sort that out and I can move on. And to Casie's point, customers need to realize that when you go through that clean-up process you're actually getting rid of a large portion of your link equity that was inflating you in the first place.
I think the baking in of the upcoming Penguin algorithm is long overdue. And it will allow us to tell clients with confidence that if you do the right things, the actions you take now will pay you back quicker than two year's time or a year's time.
I also think that if people are going to start the process of cleaning up for Penguin, they need to get it done ASAP because it's not until Google caches all the stuff that you've done that it becomes real in terms of the way Google views it.
Disavowing stuff and having it re-cached, cleaning stuff up and getting it removed, all that needs to happen well before the algo gets run or you're missing your chance. For anybody trying to do it in the last two weeks before they do the update, it's not going to work.
Question Four: What is your best tip for creating linkable content?
This discussion starts at 41:03.
And the responses wrap-up around 46:39.
Nicholas: Do your research.
Look and see if your topic really is link-worthy, see if people have written about it before, find if there are other pieces of content that have preformed well within that topic realm. Use BuzzSumo, look at the share data, look at the link data. Don't go out with a scatter-gun and just spray and pray blind content to the web, that doesn't work. Do your research, that's my number one tip.
Casie: I would say it has to be about your audience. It has to be about the people you want to share your content, the people that are actually invested. Go and see what they're already sharing, linking to, and talking about, and fill a need for them.
I love the idea of going out and finding what your audience is asking on forums and blogs, and then answering those questions for them. Take a look at what your audience is interested in, and what type of content they're already sharing and linking to, and then figure out what are the questions they have that you can answer.
Pete: Design is very important, and half the links we get are due to the execution of design.
You have to bear in mind that people typically don't like to read on the web, and this is why infographics are so successful because they're data-driven and you can execute in different visual ways.
If you have a long-form, in-depth piece of content you have to continuously keep capturing the audiences attention with large headings, large sub-headings, breaking up paragraphs, and bullet points. With long-form content most people don't actually read it, they scroll to the bottom and think it's cool and bookmark it for later.
You have to be well-trained in the psychology of design to create a linkable asset. Even if it is just a customer resource or guide, you have to be very careful as to how you lay out your content because nobody is going to link to just 600 words of text. You have to really work hard to break that down and make it more readable.
Almost everything you make has to be sort of like a beautiful PowerPoint presentation.
Paul: Start with the end in mind.
When you produce content, think about why people would share it, and then think about how they would share it. If they're going to share it via social, then aim the content towards social sharing. If they're going to share it via attribution with a link, then create the content for that purpose.
Creating great content and then going out and trying to find an audience for it is a lot harder than finding the audience and creating content to match.
Nicholas: That's a really good point, people do share and link for different reasons. A lot of the time, top of the funnel content marketing is for shares and links. But I think there is more segmentation between different types of top of the funnel content, some is more likely to attract shares versus links. Shorter, bite-sized content is probably going to be shared, long-form is more likely to be linked if it is more evergreen or resourceful.
As Pete pointed out, people often share as a bookmark. But they only link if they have read the content and digested it.
Question Five: If you're not doing well in organic search (due to potential negative SEO) and PPC is extremely competitive/expensive, what would you advise?
The responses to question five being at 46:41.
The panelists discuss this question until 52:01.
Casie: I would say work on disavowing what you can, but also find out what other channels your customers are on. While you're building up your site, look at social, forums, third-party sites, and see if there are things you can be doing elsewhere as you work to get rid of that negative SEO.
Nicholas: Investing in other marketing channels is good advice.
Paul: First, find out if it really is negative SEO. Because 90% of the cases where people suggest it's negative SEO, it isn't actually negative SEO.
If it is actually negative SEO, we (at Kerboo) have something called 'Daily Imports' where we pull in all the new links that arrive on a daily basis, and you can throw them into a disavow file as they arrive. That works pretty well to sort that problem out, once you've identified it. But you have to make sure it is negative SEO before you start, and not some other reason why you're visibility is suppressed in some way. It is rarely negative SEO in my experience.
Pete: You have to question the SEO strategy itself more than anything else, excluding the backlinks. When you compare your business, your website, and your brand to people ranking effectively, are your really good enough? Are you really better than them when it comes to every SEO element?
I think it's worth doing a full technical audit to really judge if it's negative SEO. It might not just be negative SEO that is holding you back.
Nicholas: I hate fighting against an unseen, unknown algorithm without any indication that you're really helping the situation or that all your efforts are being hindered by something you can't fix.
I'm jaded because we haven't seen a Penguin refresh in so long. So I'm of the opinion that if I'm fighting a brick wall like that, paying $400 a click, I'd rather invest that money in a new site and start fresh. If I somehow clean-up the original site, I'll 301 redirect them together and go from there. But I don't like the idea of spending $400 a click waiting a year for a Penguin to refresh. Hopefully, we see it switch to real-time and my philosophy can change.
Question Six: What are your opinions regarding the future of manual link building?
Question six starts around 52:11.
The discussion for this question ends at 54:12
Nicholas: My opinion is that you should go to EricWard.com and read his rebuttal response, and that will give you the entire rundown.
It does make me a little frustrated though, that we are constantly battling the sordid history of spam that link building came out of. But there are people like Eric Ward that have been building links for twenty years, which is incredible when you consider the history of the internet, and they've been doing it the right way the whole time. And they called it link building.
I think it's too bad that spam gets called link building. That's spam building, that is spam, that's not link building.
Question Seven: What are the best tools for checking backlinks?
The last question begins at 54:17.
And the webinar discussion concludes at 58:24.
Nicholas: Majestic is of course the biggest index out there, and Ahrefs comes close. Majestic has the largest historic database, so I really like it for link analysis and clean-up.
Of course you have Moz, they have a smaller index, but they crawl in different ways and they look at different factors. They say they are more metric-based, so that helps them get the best links even though it's a smaller index.
You also have WebMeUp, which is a pretty new free tool. They have a really big index, and I find that they sometimes find things missed by Moz or Majestic.
And of course Webmaster Tools/Google Search Console is the best free tool, although it's only a sampling of the full data Google has.
The best thing you can do is go out and get as many sources as you can, because they are all different.
Pete: Google Webmaster Tools is a great free tool, and then we also have licenses for Ahrefs and Majestic. That's it for me, that's gospel.
Nicholas: Yeah the way I see it is you should always use your Google Search Console data because it's free, and then subscribe to one of the big indexes.
Paul: As far as backlink data goes, we did some analysis of backlink data in terms of the completeness of each of the indexes, and their helpfulness in doing the job of analyzing backlink profiles.
We ranked them on a scale of 100 where Google Search Console is 100. We ranked Majestic at 75, Ahrefs at 70, Moz at 20, and WebMeUp at 15. These are however on a sample case, and if we re-ranked them it might be different.
You can get a lot of your link data for free, it's a matter of making sense of that data. You can use tools like Kerboo or URLprofiler to analyze backlink data.
Nicholas: I love URLprofiler, it's my gospel. Once I do have all the link data from all the different sources, I compile it all into one place with URLprofiler.
And that's a wrap!
It was an awesome webinar and so much fun hearing and learning from all the panelists. Thank you to everyone who attended, and thanks so much to our panelists: Casie, Paul, Pete, and Nicholas! Until next time!