Backlinks are foundational pillars of any solid SEO strategy, but not all are created equal. Powerful editorial links from sites like The New York Times, CNN, or TechCrunch can help launch your business into the public spotlight and drive some serious sales. Many backlinks aren’t worth the hyperlink they’re coded into and offer little overall support own their own. Much like opinions, backlinks from highly respected or celebrity websites tend to carry much more weight than backlinks from random websites like personal blogs. People understand influence in this way, and Google’s quest to identify value mimics our own perception in many ways. Understanding this similarity more fully can help to apply some hacks and tricks to leverage the power of opinion, and backlinks, in favor of your website and marketing campaign.
As we stroll throughout the world we are constantly exposed to new things and ideas. The human mind associates similar experiences, thoughts, and opinions to within constructs known as mental schemas. These are like labeled shoe boxes that our minds put similar objects into to keep everything better organized. You might have a schema for first introductions, job interviews, and even checking the mail. Over the course of our evolution, our minds have evolved this way—in theory—as a way to simplify cognitive processing task to reduce our overall use of energy. After all, it’d be hard to know which friend to call to help you move furniture if you had to re-evaluate what friendship meant every time you talked.
If a Senator were to recommend a hot dog stand you’d likely try it out no questions asked. If a stranger made the same recommendation you’d likely ignore it
One of the shadows of this mental organization structure is that experiences and subjects common to everyone tend to get an aggregated Schema. For example, if you’d read in every local publisher’s content that some new restaurant had the best-roasted duck in town, would you be able to tell if they’d actually served you a goose? Most people would likely just assume their own perception was amiss since their mental schema for local restaurant reviews favors accuracy, honesty, and trust. Some might call it herd mentality, but it could arguably be regarded simply as our inborn informational storage architecture. This is a loophole in human psychology that is often leveraged by less-scrupulous salesmen and could be considered the ‘bait and switch’ by many regards.
The Weight of Clout
Opinions of people are often perceived in a similar capacity—those from well-known people being given more weight. Comparatively speaking, if a US Senator and a complete stranger were to both answer your Facebook request for a local restaurant recommendation—which one would you try? Nine times of ten, the Senator’s opinion is going to win the battle. Nothing is ever so simple, and understanding the dynamics of this perception can help understand how opinion can be influenced. For example, if one Senator were to recommend a hot dog stand you’d likely try it out no questions asked. If a stranger made the same recommendation, especially if it were among a crowd of other stranger’s differing recommendations, you’d likely ignore it. Mental Schemas—simple survival psychology in action. Where things get interesting is when you start to consider quantity of opinions. If the Senator told you to eat at Joe’s Authentic Coney Island Dogs, and a stranger told you to eat at Mike’s Chicago Style Footlongs, you’re going Coney Island everytime. What if 5 strangers told you to go Chicago style? What if were 50, 100, or maybe 1000? At this point most everyone would go Chicago-style, and likely even wonder if there is some reason not to trust the Senator. This is an example of quantity vs. quality of opinion, when a mass-agreement of strangers can trump the opinion of high society or even experts in the field sometimes. Common real-world examples of this are political interests promoting false rumors, drug companies buying the opinions of peer-reviewed doctors, and brands promoting anything with celebrity endorsement.
Mental Schemas are, very simply speaking, the constructs in which we organize our intelligence of the world. When we examine the world of artificial intelligence, however, our vocabulary changes slightly. Deep learning algorithms, such as the technologies used by Google to build their search databases, use many different approaches to build relationships between different objects. One of particular note to SEO services and search engine ranking is the concept of Word Vectors. Very simply put, word vectors are data structures by which computers can store words as numerical values—which allows very in-depth comparisons to be made. For example, word vectors help tell Google that ‘Man’ is to ‘King’ in the same capacity that ‘Woman’ is to ‘Queen.’ This is one of the ways that Google knows to show Mike’s Chicago Style Hotdogs’ webpage when you search for ‘hot dogs’ instead of pictures of border collies standing in the Sahara Desert.
If tons of websites were to start talking about ‘hog dogs’ in the context of Border Collies in the Sahara Desert, Google’s word vector representations of these topics could be influenced to favor showing images of desert gods over Mike’s Hotdog Stand. Another very interesting facet about this approach of learning word relationships is that different communities regard different subjects are having different degrees of similarity. A great interactive demo of this technology in action can be seen by the SpaCy’s Sense2Vec word vector demonstration which takes into account only comments posted to Reddit. For example, Reddit has a 92% similar association between ‘bacon’ and ‘salad’. We often find this type of customized approach to be quite useful in better understanding the subtleties of targeted demographics. The more certain topics are mentioned in context with one another, the more they’ll be regarded as similar keywords by word vector approaches.
Backlinks Are Like Digital Opinions
To bring all this together, let’s consider what we know; more highly respected people lend more highly respected opinions, though large volumes of opinions can also amass strong consideration. Our mental schemas help guide our understanding of similar concepts and objects, and word vectors help guide AI’s understanding of similarity between different objects—in often comically revealing ways. The core aspect of Google’s search engine is its functionality to help people find what they consider relevant and valuable to what they searched for. Google calculates this in many different ways, but backlinks and word vectors make up two essential facets. When a thousand websites have a sentence with the keywords “hot dog” and “bun” in them, Google knows that “hot dog bun” is a related keyword, and may even offer it as a suggestion. When a thousand websites that are talking about “hot dogs,” “buns,” and “hot dog buns” also have a link to Mike’s Chicago Style Hotdogs page—Google ranks that page’s chance of providing value to a search for “hot dogs” as much stronger than that of a page with no links. In many ways, Google regards backlinks much like we regard the opinions of others. If the New York Times were to link to to a website about hotdogs, Google is going to regard that as carrying more weight than a link from a foodie’s personal WordPress blog. Just like we tend to trust the opinions of those who other’s tend to trust—Google places more weight on the backlinks from sites that tend to get more authoritative backlinks.
When there is any system of predictability, there exists an opportunity to leverage that knowledge however you may choose. If you know that your microwave always beeps and cuts off when it’s done cooking, you can walk into the other room and get your movie started while you’re waiting for your popcorn to warm up. If you didn’t know your microwave would beep and stop heating when your popcorn was ready, you’d have to stand there and monitor it the entire time to ensure it didn’t burn. This is an example of being able to accurately predict the future, and how one can leverage to their advantage. People’s perception of mass events and Google’s processing of mass quantities of data are both systems that have a degree of predictability build into them. A hundred positive reviews about the local restaurant’s duck can mask your awareness of your meal being goose in the same way that a handful of links from CNN, The New York Times, and MSNBC can sway public sentiment regarding matters of politics, economics, or even trending technology. When sources of opinion deliver consistently accurate perceptions, no one really seems to notice when they slip up—either intentionally or purposefully. When this system is applied in practice in a predictable way, restaurant owners can plan on substituting goose for duck and build it into the overall cost or the product.
Large amounts of … backlinks can still hack Google’s search algorithm the same way that large amounts of opinions from stranger’s can hack our personal opinions.
Applying this into the realm of ranking websites and potent SEO services, one needs only substitute the concept of ‘opinion’ with ‘backlink’. If you’re ranking a website and get a single backlink from a CNN or New York Times with strong contextual reference—Google is going to respect that opinion as a strong indication of your page providing value for keywords it believes are related to terms found on both pages. Where things get interesting is when you start comparing the effect of lots of average—and maybe even untrustworthy—backlinks to low amounts of trusted backlinks. For example, if Mike’s Chicago Style Hotdog’s page had a link from the Chicago Tribune—and no other links—and Joe’s Coney Island Dogs had 50 links from random food blogs, who do you think would outrank the other? Nowadays it’s all about local search, but consider if they were in the same city, maybe even the same block. Now, what about some newcomer’s hotdog stand that had 5000 backlinks from really low-quality websites? That’d be like 5000 opinions of strangers telling you that this establishment is worth your patronage—and would likely catch your attention if nothing else. Google has gotten a lot smarter in recent years an does a much better job of handling the discerning low-quality backlinks, but it can still get tripped up—just like we could if 5000 people were to start chanting in unity. Simply put; creating large amounts of highly-relevant backlinks can still ‘hack’ Google’s search algorithm the same way that large amounts of opinions from strangers can ‘hack’ our opinions.
Backlinks are, in many ways, much like digital versions of opinions. The more trustworthy sources more strongly influence SEO ranking factors measured by Google. Recognizing how Google builds associations and perceives large amounts of backlinks can help take some serious shortcuts in SEO. These principles will likely always be able to be leveraged in some capacity because in many regards Google’s search algorithm is a mirrored interpretation of how humans perceive associations and value between keywords and results. In the real world, these relationships are leveraged all the time, often in ways that resemble so-called Black Hat SEO methods like building Web 2.0 Link Pyramids. Don’t believe me? Check out this video made popular by Conan Obrien (YouTube) that mashes up the ‘opinions’ of several local news station anchors. While you’re watching, imagine them as Web 2.0 sites—all working together to create a forced similarity in order to leverage a broader range of perception. This is a great example of how real-world information marketing reflects digital information marketing, both of which serve the similar purpose of influencing an audience. SEO and digital marketing is a large arena in which many different facets are leveraged to help control opinions—and backlinks.