Over the last few years the sky, or at least the bottom line, has fallen for many websites and business owners as a result of major algorithmic changes. But that doesn’t necessarily mean pulling the rip cord on all things links.
Some recent public commentary and studies have gotten many people thinking about the future of links for SEO and whether or not their days as the preeminent ranking factor are truly numbered.
First, there was what John Mueller said last month in a hangout – which has already inspired blog posts – and plenty of Twitter conversation.
About a year ago, Matt Cutts actually divulged that Google did in fact have their own inside version of a SERPs without links as a ranking factor that they had “experimented” with.
Then, last week a new study from Google talked about “a new approach that relies on endogenous signals, namely, the correctness of factual information provided by the source.” So, fact-based rankings, rather than popularity-based rankings.
Are these horsemen of the linkpocalypse? Could these all be signs of the end of days for links?
In the SEO industry it’s easy to think the sky is falling. Because over the last couple of years the sky, or at least the bottom line, has fallen for many websites and business owners as a result of major algorithmic changes. But that doesn’t necessarily mean pulling the rip cord on all things links.
Avoiding “link-building” is not necessarily the same as avoiding “link acquisition.” It is also not the same as “don’t try to get other sites to link to you.”
I think part of the complexity here is that various terms in SEO become blanket concepts for an entire genre of activities. For instance, “link bait,” can mean any form of content that you hope will be linked to, from an infographic to an incendiary diatribe against a person or practice. The term “link-building” can mean anything from writing an e-book for a specific market and promoting it to a targeted audience, to purchasing links from the guy who emailed you at 2 a.m. on a Sunday about his directory of “PR 4 links.” Yes, sadly, that’s still a thing.
It’s possible to interpret Mr. Mueller’s statement as an indictment of low-quality link-building programs that focus entirely on the link itself without a thought of value of the content being linked to. Obviously, these are the kinds of networks and link farms that have largely been algorithmically damned by Penguin already. A cynic might say that, these days, with all the “warning” signs, if you wander into the minefield of cheap, easy links, then you kind of deserve to get blown up. But that’s not to say that all efforts to “build links” are inherently dangerous.
So while this distinction is not necessarily verbalized, he does give tips on how to make content more linkable, implying that having people ink to you is still a good, and necessary practice. But that focusing on links alone is where strategy can cross into dangerous territory.
But that does not unequivocally mean that link acquisition, even content marketing with the goal of attracting links, is necessarily inherently bad or to be avoided. In fact, both Mueller and Cutts respectively say, links are still a major part of the ranking algorithm, just not the only one.
Speaking of what Cutts had to say about testing a version of rankings that did not include link data, the existence of the experiment and its results are both telling. The fact that Google has considered removing links from the ranking equation is not surprising, given the massive changes that have been implemented with regard to how link data is processed and interpreted. Is it so far-fetched Google has at least considered a world where links don’t matter? Absolutely not.
But Cutts tells us quite clearly here that search results, without back link data, fall short in terms of quality. In the video he says, “SERPS without links as a ranking factor look even worse than they do now.” He goes onto explain that back link data, in spite of the “noise and spam” are still a “big win” in terms making sure that Google returns the “best”, “most relevant “and “most popular” results.
The existence of the tested data alone tells us that Google itself has questioned the preeminence and even validity of back link data in the ranking formula but that by and large, as it stands is still the most reliable way of ensuring its best product: quality search results.
Even though this could potentially indicate a contradiction between “avoid link-building” and “links still give the best data,” there is only a dichotomy if we choose to ignore nuance and lump all activities that result in link acquisition under the same umbrella. The basic practice of earning links as a result of quality and popularity are implicitly implied by both Google gurus as an important part of digital marketing.
From a completely separate Google source, a team of research scientists published a paper evaluating the prospects for a concept called Knowledge-Based Trust. This would create rankings that were based more on factual accuracy than link-based popularity. This methodology would rely heavily on the Knowledge Base that Google has been building and already displaying through its Knowledge Graph. The process would evaluate any given Web page for its accuracy in conjunction with other pages on the website and compared against the vault of knowledge Google has built up.
While this is yet another variation on using non-link based signals for rankings, it isn’t necessarily a replacement. The biggest weakness in this approach to rankings is that it cannot scale across all verticals. There are simply too many areas in which results cannot be based on fact and must be based on opinion. For example, there is a vast difference between searching for “Who is William Shakespeare” and “What is William Shakespeare’s Greatest Play.”
Results for Query 1 can be evaluated based on agreed about fact, and displayed in ordered relating to the estimated multi-page accuracy of a site bearing the answer to this question.
The results for Query 2, however, could be debated until the end of time. There is no fact, only opinion. In these cases isn’t it, at least arguably, better to rank the results based on a metric like links where many people have “voted” for either the validity or comprehensiveness of one opinion over another?
These are, no doubt, the kinds of questions Google must wrestle with in terms of evaluating how and how much to implement this kind of concept as an algorithmic factor. Clearly, it is not necessarily appropriate in all cases. That being the case, we can’t count on “factual accuracy” replacing crowd-based signals like links in the immediate future, and potentially not ever in many areas of search.
So while some might say recent events are a death knell for SEO-related link-building, things may not be that dramatic, at least not yet. For the foreseeable future, links still have a role to play. Links are still essential for bots to crawl and index the Web. And so far, Google has simply had too much success in creating a SERP product of reasonable quality using links as a dominant factor. That’s not to say that the proportions within the equation are not, and will not continue to shift. Certainly greater weight can and will be given to additional factors (think mobile accessibility) as we go forward. And while actively link-building may make you feel as though you’re living under Damocles’ Sword, you should be able to rest easy as long as you are doing it with an emphasis on content and authenticity.