How Much Has Link Building Changed in Recent Years?

Posted by Paddy_Moogan

I get asked this question a lot. It’s mainly asked by people who are considering buying my link building book and want to know whether it’s still up to date. This is understandable given that the first edition was published in February 2013 and our industry has a deserved reputation for always changing.

I find myself giving the same answer, even though I’ve been asked it probably dozens of times in the last two years—”not that much”. I don’t think this is solely due to the book itself standing the test of time, although I’ll happily take a bit of credit for that 🙂 I think it’s more a sign of our industry as a whole not changing as much as we’d like to think.

I started to question myself and if I was right and honestly, it’s one of the reasons it has taken me over two years to release the second edition of the book.

So I posed this question to a group of friends not so long ago, some via email and some via a Facebook group. I was expecting to be called out by many of them because my position was that in reality, it hasn’t actually changed that much. The thing is, many of them agreed and the conversations ended with a pretty long thread with lots of insights. In this post, I’d like to share some of them, share what my position is and talk about what actually has changed.

My personal view

Link building hasn’t changed as much we think it has.

The core principles of link building haven’t changed. The signals around link building have changed, but mainly around new machine learning developments that have indirectly affected what we do. One thing that has definitely changed is the mindset of SEOs (and now clients) towards link building.

I think the last big change to link building came in April 2012 when Penguin rolled out. This genuinely did change our industry and put to bed a few techniques that should never have worked so well in the first place.

Since then, we’ve seen some things change, but the core principles haven’t changed if you want to build a business that will be around for years to come and not run the risk of being hit by a link related Google update. For me, these principles are quite simple:

  • You need to deserve links – either an asset you create or your product
  • You need to put this asset in front of a relevant audience who have the ability to share it
  • You need consistency – one new asset every year is unlikely to cut it
  • Anything that scales is at risk

For me, the move towards user data driving search results + machine learning has been the biggest change we’ve seen in recent years and it’s still going.

Let’s dive a bit deeper into all of this and I’ll talk about how this relates to link building.

The typical mindset for building links has changed

I think that most SEOs are coming round to the idea that you can’t get away with building low quality links any more, not if you want to build a sustainable, long-term business. Spammy link building still works in the short-term and I think it always will, but it’s much harder than it used to be to sustain websites that are built on spam. The approach is more “churn and burn” and spammers are happy to churn through lots of domains and just make a small profit on each one before moving onto another.

For everyone else, it’s all about the long-term and not putting client websites at risk.

This has led to many SEOs embracing different forms of link building and generally starting to use content as an asset when it comes to attracting links. A big part of me feels that it was actually Penguin in 2012 that drove the rise of content marketing amongst SEOs, but that’s a post for another day…! For today though, this goes some way towards explain the trend we see below.

Slowly but surely, I’m seeing clients come to my company already knowing that low quality link building isn’t what they want. It’s taken a few years after Penguin for it to filter down to client / business owner level, but it’s definitely happening. This is a good thing but unfortunately, the main reason for this is that most of them have been burnt in the past by SEO companies who have built low quality links without giving thought to building good quality ones too.

I have no doubt that it’s this change in mindset which has led to trends like this:

The thing is, I don’t think this was by choice.

Let’s be honest. A lot of us used the kind of link building tactics that Google no longer like because they worked. I don’t think many SEOs were under the illusion that it was genuinely high quality stuff, but it worked and it was far less risky to do than it is today. Unless you were super-spammy, the low-quality links just worked.

Fast forward to a post-Penguin world, things are far more risky. For me, it’s because of this that we see the trends like the above. As an industry, we had the easiest link building methods taken away from us and we’re left with fewer options. One of the main options is content marketing which, if you do it right, can lead to good quality links and importantly, the types of links you won’t be removing in the future. Get it wrong and you’ll lose budget and lose the trust if your boss or client in the power of content when it comes to link building.

There are still plenty of other methods to build links and sometimes we can forget this. Just look at this epic list from Jon Cooper. Even with this many tactics still available to us, it’s hard work. Way harder than it used to be.

My summary here is that as an industry, our mindset has shifted but it certainly wasn’t a voluntary shift. If the tactics that Penguin targeted still worked today, we’d still be using them.

A few other opinions…

I definitely think too many people want the next easy win. As someone surfing the edge of what Google is bringing our way, here’s my general take—SEO, in broad strokes, is changing a lot, *but* any given change is more and more niche and impacts fewer people. What we’re seeing isn’t radical, sweeping changes that impact everyone, but a sort of modularization of SEO, where we each have to be aware of what impacts our given industries, verticals, etc.”

Dr. Pete

 

I don’t feel that techniques for acquiring links have changed that much. You can either earn them through content and outreach or you can just buy them. What has changed is the awareness of “link building” outside of the SEO community. This makes link building / content marketing much harder when pitching to journalists and even more difficult when pitching to bloggers.

“Link building has to be more integrated with other channels and struggles to work in its own environment unless supported by brand, PR and social. Having other channels supporting your link development efforts also creates greater search signals and more opportunity to reach a bigger audience which will drive a greater ROI.

Carl Hendy

 

SEO has grown up in terms of more mature staff and SEOs becoming more ingrained into businesses so there is a smarter (less pressure) approach. At the same time, SEO has become more integrated into marketing and has made marketing teams and decision makers more intelligent in strategies and not pushing for the quick win. I’m also seeing that companies who used to rely on SEO and building links have gone through IPOs and the need to build 1000s of links per quarter has rightly reduced.

Danny Denhard

Signals that surround link building have changed

There is no question about this one in my mind. I actually wrote about this last year in my previous blog post where I talked about signals such as anchor text and deep links changing over time.

Many of the people I asked felt the same, here are some quotes from them, split out by the types of signal.

Domain level link metrics

I think domain level links have become increasingly important compared with page level factors, i.e. you can get a whole site ranking well off the back of one insanely strong page, even with sub-optimal PageRank flow from that page to the rest of the site.

Phil Nottingham

I’d agree with Phil here and this is what I was getting at in my previous post on how I feel “deep links” will matter less over time. It’s not just about domain level links here, it’s just as much about the additional signals available for Google to use (more on that later).

Anchor text

I’ve never liked anchor text as a link signal. I mean, who actually uses exact match commercial keywords as anchor text on the web?

SEOs. 🙂

Sure there will be natural links like this, but honestly, I struggle with the idea that it took Google so long to start turning down the dial on commercial anchor text as a ranking signal. They are starting to turn it down though, slowly but surely. Don’t get me wrong, it still matters and it still works. But like pure link spam, the barrier is a lot more lower now in terms what of constitutes too much.

Rand feels that they matter more than we’d expect and I’d mostly agree with this statement:

Exact match anchor text links still have more power than you’d expect—I think Google still hasn’t perfectly sorted what is “brand” or “branded query” from generics (i.e. they want to start ranking a new startup like meldhome.com for “Meld” if the site/brand gets popular, but they can’t quite tell the difference between that and https://moz.com/learn/seo/redirection getting a few manipulative links that say “redirect”)

Rand Fishkin

What I do struggle with though, is that Google still haven’t figured this out and that short-term, commercial anchor text spam is still so effective. Even for a short burst of time.

I don’t think link building as a concept has changed loads—but I think links as a signal have, mainly because of filters and penalties but I don’t see anywhere near the same level of impact from coverage anymore, even against 18 months ago.

Paul Rogers

New signals have been introduced

It isn’t just about established signals changing though, there are new signals too and I personally feel that this is where we’ve seen the most change in Google algorithms in recent years—going all the way back to Panda in 2011.

With Panda, we saw a new level of machine learning where it almost felt like Google had found a way of incorporating human reaction / feelings into their algorithms. They could then run this against a website and answer questions like the ones included in this post. Things such as:

  • “Would you be comfortable giving your credit card information to this site?”
  • “Does this article contain insightful analysis or interesting information that is beyond obvious?”
  • “Are the pages produced with great care and attention to detail vs. less attention to detail?”

It is a touch scary that Google was able to run machine learning against answers to questions like this and write an algorithm to predict the answers for any given page on the web. They have though and this was four years ago now.

Since then, they’ve made various moves to utilize machine learning and AI to build out new products and improve their search results. For me, this was one of the biggest and went pretty unnoticed by our industry. Well, until Hummingbird came along I feel pretty sure that we have Ray Kurzweil to thank for at least some of that.

There seems to be more weight on theme/topic related to sites, though it’s hard to tell if this is mostly link based or more user/usage data based. Google is doing a good job of ranking sites and pages that don’t earn the most links but do provide the most relevant/best answer. I have a feeling they use some combination of signals to say “people who perform searches like this seem to eventually wind up on this website—let’s rank it.” One of my favorite examples is the Audubon Society ranking for all sorts of birding-related searches with very poor keyword targeting, not great links, etc. I think user behavior patterns are stronger in the algo than they’ve ever been.

– Rand Fishkin

Leading on from what Rand has said, it’s becoming more and more common to see search results that just don’t make sense if you look at the link metrics—but are a good result.

For me, the move towards user data driving search results + machine learning advanced has been the biggest change we’ve seen in recent years and it’s still going.

Edit: since drafting this post, Tom Anthony released this excellent blog post on his views on the future of search and the shift to data-driven results. I’d recommend reading that as it approaches this whole area from a different perspective and I feel that an off-shoot of what Tom is talking about is the impact on link building.

You may be asking at this point, what does machine learning have to do with link building?

Everything. Because as strong as links are as a ranking signal, Google want more signals and user signals are far, far harder to manipulate than established link signals. Yes it can be done—I’ve seen it happen. There have even been a few public tests done. But it’s very hard to scale and I’d venture a guess that only the top 1% of spammers are capable of doing it, let alone maintaining it for a long period of time. When I think about the process for manipulation here, I actually think we go a step beyond spammers towards hackers and more cut and dry illegal activity.

For link building, this means that traditional methods of manipulating signals are going to become less and less effective as these user signals become stronger. For us as link builders, it means we can’t keep searching for that silver bullet or the next method of scaling link building just for an easy win. The fact is that scalable link building is always going to be at risk from penalization from Google—I don’t really want to live a life where I’m always worried about my clients being hit by the next update. Even if Google doesn’t catch up with a certain method, machine learning and user data mean that these methods may naturally become less effective and cost efficient over time.

There are of course other things such as social signals that have come into play. I certainly don’t feel like these are a strong ranking factor yet, but with deals like this one between Google and Twitter being signed, I wouldn’t be surprised if that ever-growing dataset is used at some point in organic results. The one advantage that Twitter has over Google is it’s breaking news freshness. Twitter is still way quicker at breaking news than Google is—140 characters in a tweet is far quicker than Google News! Google know this which is why I feel they’ve pulled this partnership back into existence after a couple of years apart.

There is another important point to remember here and it’s nicely summarised by Dr. Pete:

At the same time, as new signals are introduced, these are layers not replacements. People hear social signals or user signals or authorship and want it to be the link-killer, because they already fucked up link-building, but these are just layers on top of on-page and links and all of the other layers. As each layer is added, it can verify the layers that came before it and what you need isn’t the magic signal but a combination of signals that generally matches what Google expects to see from real, strong entities. So, links still matter, but they matter in concert with other things, which basically means it’s getting more complicated and, frankly, a bit harder. Of course, on one wants to hear that.”

– Dr. Pete

The core principles have not changed

This is the crux of everything for me. With all the changes listed above, the key is that the core principles around link building haven’t changed. I could even argue that Penguin didn’t change the core principles because the techniques that Penguin targeted should never have worked in the first place. I won’t argue this too much though because even Google advised website owners to build directory links at one time.

You need an asset

You need to give someone a reason to link to you. Many won’t do it out of the goodness of their heart! One of the most effective ways to do this is to develop a content asset and use this as your reason to make people care. Once you’ve made someone care, they’re more likely to share the content or link to it from somewhere.

You need to promote that asset to the right audience

I really dislike the stance that some marketers take when it comes to content promotion—build great content and links will come.

No. Sorry but for the vast majority of us, that’s simply not true. The exceptions are people that sky dive from space or have huge existing audiences to leverage.

You simply have to spend time promoting your content or your asset for it to get shares and links. It is hard work and sometimes you can spend a long time on it and get little return, but it’s important to keep working at until you’re at a point where you have two things:

  • A big enough audience where you can almost guarantee at least some traffic to your new content along with some shares
  • Enough strong relationships with relevant websites who you can speak to when new content is published and stand a good chance of them linking to it

Getting to this point is hard—but that’s kind of the point. There are various hacks you can use along the way but it will take time to get right.

You need consistency

Leading on from the previous point. It takes time and hard work to get links to your content—the types of links that stand the test of time and you’re not going to be removing in 12 months time anyway! This means that you need to keep pushing content out and getting better each and every time. This isn’t to say you should just churn content out for the sake of it, far from it. I am saying that with each piece of content you create, you will learn to do at least one thing better the next time. Try to give yourself the leverage to do this.

Anything scalable is at risk

Scalable link building is exactly what Google has been trying to crack down on for the last few years. Penguin was the biggest move and hit some of the most scalable tactics we had at our disposal. When you scale something, you often lose some level of quality, which is exactly what Google doesn’t want when it comes to links. If you’re still relying on tactics that could fall into the scalable category, I think you need to be very careful and just look at the trend in the types of links Google has been penalizing to understand why.

The part Google plays in this

To finish up, I want to briefly talk about the part that Google plays in all of this and shaping the future they want for the web.

I’ve always tried to steer clear of arguments involving the idea that Google is actively pushing FUD into the community. I’ve preferred to concentrate more on things I can actually influence and change with my clients rather than what Google is telling us all to do.

However, for the purposes of this post, I want to talk about it.

General paranoia has increased. My bet is there are some companies out there carrying out zero specific linkbuilding activity through worry.

Dan Barker

Dan’s point is a very fair one and just a day or two after reading this in an email, I came across a page related to a client’s target audience that said:

“We are not publishing guest posts on SITE NAME any more. All previous guest posts are now deleted. For more information, see www.mattcutts.com/blog/guest-blogging/“.

I’ve reworded this as to not reveal the name of the site, but you get the point.

This is silly. Honestly, so silly. They are a good site, publish good content, and had good editorial standards. Yet they have ignored all of their own policies, hard work, and objectives to follow a blog post from Matt. I’m 100% confident that it wasn’t sites like this one that Matt was talking about in this blog post.

This is, of course, from the publishers’ angle rather than the link builders’ angle, but it does go to show the effect that statements from Google can have. Google know this so it does make sense for them to push out messages that make their jobs easier and suit their own objectives—why wouldn’t they? In a similar way, what did they do when they were struggling to classify at scale which links are bad vs. good and they didn’t have a big enough web spam team? They got us to do it for them 🙂

I’m mostly joking here, but you see the point.

The most recent infamous mobilegeddon update, discussed here by Dr. Pete is another example of Google pushing out messages that ultimately scared a lot of people into action. Although to be fair, I think that despite the apparent small impact so far, the broad message from Google is a very serious one.

Because of this, I think we need to remember that Google does have their own agenda and many shareholders to keep happy. I’m not in the camp of believing everything that Google puts out is FUD, but I’m much more sensitive and questioning of the messages now than I’ve ever been.

What do you think? I’d love to hear your feedback and thoughts in the comments.

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Reblogged 4 years ago from tracking.feedpress.it

A Universal SEO Strategy Audit in 5 Steps – Whiteboard Friday

Posted by randfish

When it comes to building an SEO strategy, many marketers (especially those who don’t spend a significant amount of time with SEO) start off by asking a few key questions. That’s a good start, but only if you’re asking the right questions. In today’s Whiteboard Friday, Rand puts the usual suspects on the chopping block, showing us the five things we should really be looking into when formulating our SEO strategy.

For reference, here’s a still of this week’s whiteboard!

Video transcription

Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we’re chatting about building an SEO strategy and having a universal set of five questions that can get you there.

So number one: What keywords do you want to rank for?


Number two
: How do we get links?


Number three
: Site speed. Mobile? Doesn’t even seem like a question.


Number four
: What about Penguin and Panda?


Number five
: When do I get money?

This is bologna. That’s not a strategy. Some of those go to tactics you might invest in an SEO, but this is not an SEO strategy. Unfortunately, this is how a lot of conversations about SEO start at teams, with CMOs, with managers, with CEOs, with clients or potential clients, and it’s very frustrating because you can’t truly do a great job with SEO just in the tactical level. If you don’t start with a compelling strategy, doing all of these things is only going to produce a small amount of potential return compared to if you ask the right questions and you get your strategy set before you begin an SEO process and nailing your tactics.

So that’s what I want to go through. I spend a lot of time thinking through these things and analyzing a lot of posts that other people have put up and questions that folks have put in our Q&A system and others, on Quora and other places. I think actually every great SEO strategy that I have ever seen can be the distilled down to answers that come from these five questions.

So number one: What does our organization create that helps solve searchers’ questions or problems? That could be, “Or what will we create in the future?” It might be that you haven’t yet created the thing or things that’s going to help solve searchers’ questions or problems. But that thing that you make, that product or service or content that you are making, that expertise that you hold, something about your organization is creating value that if only searchers could access it, they would be immensely thankful.

It is possible, and I have seen plenty of examples of companies that are so new or so much on the cutting edge that they’re producing things that aren’t solving questions people are asking yet. The problem that you’re solving then is not a question. It’s not something that’s being searched for directly. It usually is very indirect. If you’re creating a machine that, let’s say, turns children’s laughter into energy, as they do in the film “Monsters, Inc.”, that is something very new. No one is searching for machine to turn kids laughing into energy. However, many people are searching for alternative energy. They’re searching for broader types of things and concepts. By the way, if you do invent that machine, I think you’ll probably nail a lot of that interest level stuff.

If you have a great answer to this, you can then move on to, “What is the unique value we provide that no one else does?” We talked about unique value previously on Whiteboard Friday. There’s a whole episode you can watch about that. Basically, if everyone else out there is producing X and X+1 and X+2, you’ve either got to be producing X times 10, or you’ve got to be producing Y, something that is highly unique or is unique because it is of such better quality, such greater quality. It does the job so much better than anything else out there. It’s not, “Hey, we’re better than the top ten search results.” It’s, “Why are you ten times better than anything on this list?”

The third question is, “Who’s going to help amplify our message, and why will they do it?” This is essential because SEO has turned from an exercise, where we essentially take content that already exists or create some content that will solve a searcher problem and then try and acquire links to it, or point links to it, or point ranking signals at it, and instead it’s ones where we have to go out and earn those ranking signals. Because we’ve shifted from link building or ranking signal building to ranking signal earning, we better have people who will help amplify our message, the content that we create, the value that we provide, the service or the product, the message about our brand.

If we don’t have those people who, for some reason, care enough about what we’re doing to help share it with others, we’re going to be shouting into a void. We’re going to get no return on the investment of broadcasting our message or reaching out one to one, or sharing on social media, or distributing. It’s not going to work. We need that amplification. There must be some of it, and because we need amplification in order to earn these ranking signals, we need an answer to who.

That who is going to depend highly on your target audience, your target customers, and who influences your target customers, which may be a very different group than other customers just like them. There are plenty of businesses in industries where your customers will be your worst amplifiers because they love you and they don’t want to share you with anyone else. They love whatever product or service you’re providing, and they want to keep you all to themselves. By the way, they’re not on social media, and they don’t do sharing. So you need another level above them. You need press or bloggers or social media sharers, somebody who influences your target audience.

Number four: What is our process for turning visitors from search into customers? If you have no answer to this, you can’t expect to earn search visits and have a positive return on your investment. You’ve got to be building out that funnel that says, “Aha, people have come to us through channel X, search, social media, e-mail, directly visited, referred from some other website, through business development, through conference or trade show, whatever it is. Then they come back to our website. Then they sign up for an e-mail. Then they make a conversion. How does that work? What does our web-marketing funnel look like? How do we take people that visited our site for the first time from search, from a problem or a question that they had that we answered, and now how do they become a customer?” If you don’t have that process yet, you must build it. That’s part of a great SEO strategy. Then optimization of this is often called conversion rate optimization.

The last question, number five: How do we expose what we do that provides value here in a way that engines can easily crawl, index, understand, and show off? This is getting to much more classic SEO stuff. For many companies they have something wonderful that they’ve built, but it’s just a mobile app or a web app that has no physical URL structure that anyone can crawl and be exposed to, or it’s a service based business.

Let’s say it’s legal services firm. How are we going to turn the expertise of our legal team into something that engines can perceive? Maybe we have the answers to these questions, but we need to find some way to show it off, and that’s where content creation comes into play. So we don’t just need content that is good quality content that can be crawled and indexed. It also must be understood, and this ties a little bit to things we’ve talked about in the past around Hummingbird, where it’s clear that the content is on the topic and that it really answers the searchers’ underlying question, not just uses the keywords the searcher is using. Although, using the keywords is still important from a classic SEO perspective.

Then show off that content is, “How do we do a great job of applying rich snippets, of applying schema, of having a very compelling title and description and URL, of getting that ranked highly, of learning what our competitors are doing that we can uniquely differentiate from them in the search results themselves so that we can improve our click-through rates,” all of those kinds of things.

If you answer these five questions, or if your customer, your client, your team, your boss already has great answers to these five questions, then you can start getting pretty tactical and be very successful. If you don’t have answers to these yet, go get them. Make them explicit, not just implicit. Don’t just assume you know what they are. Have them list them. Make sure everyone on the team, everyone in the SEO process has bought into, “Yes, these are the answers to those five questions that we have. Now, let’s go do our tactics.” I think you’ll find you’re far more successful with any type of SEO project or investment.

All right gang, thanks so much for joining us on Whiteboard Friday, and we’ll see you again next week. Take care.

Video transcription by Speechpad.com

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Reblogged 4 years ago from tracking.feedpress.it

The Future of Link Building

Posted by Paddy_Moogan

Building the types of links that help grow your online business and organic search traffic is getting harder. It used to be fairly straightforward, back before Google worked out how to treat links with different levels of quality and trust. However, the fact that it’s getting harder doesn’t mean that it’s dead.

What does the future hold?

I’m going to talk about links, but the truth is, the future isn’t really about the links. It is far bigger than that.

Quick sidenote: I’m aware that doing a blog post about the future of link building the week of a likely Penguin update could leave me with egg on my face! But we’ll see what happens.

Links will always be a ranking factor in some form or another. I can see the dials being turned down or off on certain aspects of links (more on that below) but I think they will always be there. Google is always looking for more data, more signals, more indicators of whether or not a certain page is a good result for a user at a certain moment in time. They will find them too, as we can see from
patents such as this. A natural consequence is that other signals may be diluted or even replaced as Google becomes smarter and understands the web and users a lot better.

What this means for the future is that the links valued by Google will be the ones you get as a result of having a great product and great marketing. Essentially, links will be symptomatic of amazing marketing. Hat tip to
Jess Champion who I’ve borrowed this term from.

This isn’t easy, but it shouldn’t be. That’s the point.

To go a bit further, I think we also need to think about the bigger picture. In the grand scheme of things, there are so many more signals that Google can use which, as marketers, we need to understand and use to our advantage. Google is changing and we can’t bury our heads in the sand and ignore what is going on.

A quick side note on spammy links

My background is a spammy one so I can’t help but address this quickly. Spam will continue to work for short-term hits and churn and burn websites. I’ve talked before about 
my position on this so I won’t go into too much more detail here. I will say though that those people who are in the top 1% of spammers will continue to make money, but even for them, it will be hard to maintain over a long period of time.

Let’s move onto some more of the detail around my view of the future by first looking at the past and present.

What we’ve seen in the past

Google didn’t understand links.

The fundamental issue that Google had for a long, long time was that they didn’t understand enough about links. They didn’t understand things such as:

  • How much to trust a link
  • Whether a link was truly editorially given or not
  • Whether a link was paid for or not
  • If a link was genuinely high quality (PageRank isn’t perfect)
  • How relevant a link was

Whilst they still have work to do on all of these, they have gotten much better in recent years. At one time, a link was a link and it was pretty much a case of whoever had the most links, won. I think that for a long time, Google was trying very hard to understand links and find which ones were high quality, but there was so much noise that it was very difficult. I think that eventually they realised that they had to attack the problem from a different angle and 
Penguin came along. So instead of focusing on finding the “good” signals of links, they focused on finding the “bad” signals and started to take action on them. This didn’t fix everything, but it did enough to shock our industry into moving away from certain tactics and therefore, has probably helped reduce a lot of the noise that Google was seeing.

What we’re seeing right now

Google is understanding more about language.

Google is getting better at understanding everything.
Hummingbird was just the start of what Google hopes to achieve on this front and it stands to reason that the same kind of technology that helps the following query work, will also help Google understand links better.

Not many people in the search industry said much when
Google hired this guy back in 2012. We can be pretty sure that it’s partly down to his work that we’re seeing the type of understanding of language that we are. His work has only just begun, though, and I think we’ll see more queries like the one above that just shouldn’t work, but they do. I also think we’ll see more instances of Googlers not knowing why something ranks where it does.

Google is understanding more about people.

I talk about this a little more below but to quickly summarise here, Google is learning more about us all the time. It can seem creepy, but the fact is that Google wants as much data as possible from us so that they can serve more relevant search results—and advertising of course. They are understanding more that the keywords we type into Google may not actually be what we want to find, nor are those keywords enough to find what we really want. Google needs more context.

Tom Anthony has
talked about this extensively so I won’t go into loads more detail. But to bring it back to link building, it is important to be aware of this because it means that there are more and more signals that could mean the dial on links gets turned down a bit more.

Some predictions about the future

I want to make a few things more concrete about my view of the future for link building, so let’s look at a few specifics.

1. Anchor text will matter less and less

Anchor text as a ranking signal was always something that works well in theory but not in reality. Even in my early days of link building, I couldn’t understand why Google put so much weight behind this one signal. My main reason for this view was that using exact match keywords in a link was not natural for most webmasters. I’d go as far as to say the only people who used it were SEOs!

I’m don’t think we’re at a point yet where anchor text as a ranking signal is dead and it will take some more time for Google to turn down the dial. But we definitely are at a point where you can get hurt pretty badly if you have too much commercial anchor text in your link profile. It just isn’t natural.

In the future, Google won’t need this signal. They will be much better at understanding the content of a page and importantly, the context of a page.

2. Deep linking will matter less and less

I was on the fence about this one for a long time but the more I think about it, the more I can see this happening. I’ll explain my view here by using an example.

Let’s imagine you’re an eCommerce website and you sell laptops. Obviously each laptop you sell will have its own product page and if you sell different types, you’ll probably have category pages too. With a products like laptops, chances are that other retailers sell the same ones with the same specifications and probably have very similar looking pages to yours. How does Google know which one to rank better than others?

Links to these product pages can work fine but in my opinion, is a bit of a crude way of working it out. I think that Google will get better at understanding the subtle differences in queries from users which will naturally mean that deep links to these laptop pages will be one of many signals they can use.

Take these queries:


“laptop reviews”

Context: I want to buy a laptop but I don’t know which one.


“asus laptop reviews”

Context: I like the sound of Asus, I want to read more about their laptops.


“sony laptop reviews”

Context: I also like the sound of Sony, I want to read more about their laptops.


“sony vs asus laptop”

Context: I’m confused, they both sound the same so I want a direct comparison to help me decide.


“asus laptop”

Context: I want an Asus laptop.

You can see how the mindset of the user has changed over time and we can easily imagine how the search results will have changed to reflect this. Google already understand this. There are other signals coming into play here too though, what about these bits of additional information that Google can gather about us:

  • Location: I’m on a bus in London, I may not want to buy a £1,000 laptop right now but I’ll happily research them.
  • Device: I’m on my iPhone 6, I may not want to input credit card details into it and I worry that the website I’m using won’t work well on a small screen.
  • Search history: I’ve searched for laptops before and visited several retailers, but I keep going back to the same one as I’ve ordered from them before.

These are just a few that are easy for us to imagine Google using. There are loads more that Google could look at, not to mention signals from the retailers themselves such as secure websites, user feedback, 3rd party reviews, trust signals etc.

When you start adding all of these signals together, it’s pretty easy to see why links to a specific product page may not be the strongest signal for Google to use when determining rankings.

Smaller companies will be able to compete more.

One of the things I loved about SEO when I first got into it was the fact that organic search felt like a level playing field. I knew that with the right work, I could beat massive companies in the search results and not have to spend a fortune doing it. Suffice to say, things have changed quite a bit now and there are some industries where you stand pretty much zero chance of competing unless you have a very big budget to spend and a great product.

I think we will see a shift back in the other direction and smaller companies with fewer links will be able to rank for certain types of queries with a certain type of context. As explained above, context is key and allows Google to serve up search results that meet the context of the user. This means that massive brands are not always going to be the right answer for users and Google have to get better at understanding this. Whether a company is classified as a “brand” or not can be subjective. My local craft beer shop in London is the only one in the world and if you were to ask 100 people if they’d heard of it, they’d all probably say no. But it’s a brand to me because I love their products, their staff are knowledgeable and helpful, their marketing is cool and I’d always recommend them.

Sometimes, showing the website of this shop above bigger brands in search results is the right thing to do for a user. Google need lots of additional signals beyond “branding” and links in order to do this but I think they will get them.

What all of this means for us

Predicting the future is hard, knowing what to do about it is pretty hard too! But here are some things that I think we should be doing.

  1. Ask really hard questions
    Marketing is hard. If you or your client wants to compete and win customers, then you need to be prepared to ask really hard questions about the company. Here are just a few that I’ve found difficult when talking to clients:

    • Why does the company exist? (A good answer has nothing to do with making money)
    • Why do you deserve to rank well in Google?
    • What makes you different to your competitors?
    • If you disappeared from Google tomorrow, would anyone notice?
    • Why do you deserve to be linked to?
    • What value do you provide for users?

    The answers to these won’t always give you that silver bullet, but they can provoke conversations that make the client look inwardly and at why they should deserve links and customers. These questions are hard to answer, but again, that’s the point.

  2. Stop looking for scalable link building tactics

    Seriously, just stop. Anything that can be scaled tends to lose quality and anything that scales is likely to be targeted by the Google webspam team at some point. A
    recent piece of content we did at Distilled has so far generated links from over 700 root domains—we did NOT send 700 outreach emails! This piece took on a life of its own and generated those links after some promotion by us, but at no point did we worry about scaling outreach for it.

  3. Start focusing on doing marketing that users love

    I’m not talking necessarily about you doing the next
    Volvo ad or to be the next Old Spice guy. If you can then great, but these are out of reach for most of us.That doesn’t mean you can’t do marketing that people love. I often look at companies like Brewdog and Hawksmoor who do great marketing around their products but in a way that has personality and appeal. They don’t have to spend millions of dollars on celebrities or TV advertising because they have a great product and a fun marketing message. They have value to add which is the key, they don’t need to worry about link building because they get them naturally by doing cool stuff.

    Whilst I know that “doing cool stuff” isn’t particularly actionable, I still think it’s fair to say that marketing needs to be loved. In order to do marketing that people love, you need to have some fun and focus on adding value.

  4. Don’t bury your head in the sand

    The worst thing you can do is ignore the trends and changes taking place. Google is changing, user expectations and behaviours are changing, our industry is changing. As an industry, we’ve adapted very well over the last few years. We have to keep doing this if we’re going to survive.

    Going back to link building, you need to accept that this stuff is really hard and building the types of links that Google value is hard.

In summary

Links aren’t going anywhere. But the world is changing and we have to focus on what truly matters: marketing great products and building a loyal audience. 

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!

Reblogged 5 years ago from feedproxy.google.com