By ANDREW WOODWARD
It’s not just about Fred…
This is part one of new ongoing blog series centered on digital marketing. To better highlight our new services, Andrew Woodward, our digital marketing specialist, will be helping you understand and better utilize industry trends, search engine updates, and best practices. This series is intended to help you see the benefits and impact of digital marketing on your overall marketing strategy.
The days of out thinking Google, and of artificially inflating page rankings are long gone. Keyword stuffing, paying for backlinks, copy and pasting content, and the buggy whip are also considered outdated and obsolete.
The latest “Fred” update, released in March, caused quite a stir in the SEO world. Overnight, without warning, some sites saw a several hundred percent increase in traffic, while others lost up to 90%. The results were all over the map.
I’m fortunate. I work at an agency where I can simply pull up analytics from a variety of client sites and see, at a glance, how a particular update affects the overall landscape. We saw major changes from the “Fred” update. Every single website was affected. The graphic below clearly shows the impact that Fred had on just three of the many websites we looked at:
It can be pretty entertaining observing all of the speculation that occurs in the days and weeks that follow a major Google update. The “Fred” update was no exception. Here are some of the theories that have been floating around in the SEO community:
• Fred targets ad heavy, low-value content
• Fred targets ad and affiliate sites
• Fred looks for “spammy” sites
• Fred targets “black hat” SEO practices
• Fred penalizes duplicate content
• Fred penalizes sites with ads that take up too much screen real estate
They probably all contain a grain of truth. Algorithm updates cannot be defined so simply as to say “it does this or that.” It’s simply not that linear. The name of the game now is not keyword matching, it’s “semantic relationships.” For the Google algorithms to accurately determine what a searcher is looking for requires analyzing the semantic relationship between the search words. At the same time, Google crawlers are looking for semantically related content on website. The more “cohesive” a site is, the better it will rank. Obviously that’s over simplifying it, but hopefully you get the idea.
Remember, too, that Google knows your search history. It knows a lot about you. For example, when you search for the phrase “windows” Google will determine, based on your previous online activity, whether you’re a carpenter seeking windows for your latest project, or you own a Windows-based computer, and are looking for the latest OS update (or something similar).
See and control your search activity
Don’t Worry -We Can Learn from This!
In this article, instead of focusing just on the impact of Fred, we’re going to share with you some of our SEO “best practices.” It’s been our experience that aligning your SEO strategy with Google’s long-term intentions will minimize the impact that future Google updates will have on your overall visibility, traffic, ranking, and authority. We’re not going to go too deep, but these SEO and design rules are really essential, regardless of what you’re trying to accomplish online. Think of it as the “if you can’t beat ‘em, join ‘em” strategy for SEO.
There are two kinds of Google updates: Those that came before RankBrain (2015) and those that came after RankBrain.
RankBrain was, and continues to be, a game changer. It is a machine-learning is to look at the semantic relationships between everything online. Contrary to popular belief, it doesn’t actually learn on its own. What it does do, however, is to think about information in the same way human beings do, which is highly based on contextual relationships.
You’ve undoubtedly heard of the historical updates, such as “Penguin,” “Payday,” “Possum,” “Panda,” and “Hummingbird.” RankBrain is simply one component of Google’s overall search algorithms, but it’s enormously important, and becoming more so every day.
Google has stated openly that Gary Illyes of Google was recently quoted as saying “RankBrain understands the conversational manner of search and can give results accordingly. It’s less about understanding the query and more about understanding how best to score the results.”
RankBrain is now used for almost every single one of the two trillion searches it processes annually. RankBrain is described as a “search refinement tool.”
What this means for the SEO world is that you cannot specifically optimize for RankBrain. It’s too smart for that. Websites (and their content) are now viewed holistically by Google. There is no more separating certain elements of a website as being more important than others. It’s all important.
As far as we can tell, the latest “Fred” update was primarily an update to RankBrain, and has begun penalizing sites that lack the kind of “human-ness” that RankBrain prefers. Basically, if Google cannot crawl the site and determine exactly what kind of purpose it serves in the real world.
Google no longer assumes that you’re always looking to purchase something. For example, if you type in the search phrase “Ford motor company,” what you’ll get are a slew of results that all speak to the company itself, not necessarily the products. Here is a list of the results:
Search phrase: “ford motor company”:
It wasn’t too long ago that the ATF (above-the-fold) results for such a phrase would have resulted in a list comprised of dealerships and sales-based sites that were aggressively competing for top SERP positions. Now, however, Google knows what I’m looking for – and it’s not to buy a new car. Not only is it because there is nothing in my recent search history that has anything to do with searching for a new car; it is also because of the phrasing. Typing in “Ford motor company” is not something you do when you’re looking to buy a new Ford. Google knows that. However, if you alter the search phrase just slightly, an entirely different set of results is displayed.
Search phrase: “buy ford allentown pa”:
In this case, we see four paid ads from local car dealerships, plus:
Google knows I’m trying to find out where to buy a car in my area.
Then, by combining words from both searches, we see another list of search results altogether.
Search phrase: “buy ford motor company”:
Now, combining “buy” with the formal phrase “ford motor company,” Google surmises that I’m trying to buy stock in the “Ford Motor Company.”
That’s a very basic example, but it shows how LSI (Latent Semantic Indexing) works. It seeks to derive meaning by considering all of the words used, as well as their specific positions in a search query, then match the searcher with the most appropriate information. It doesn’t always get it right, but it’s amazingly accurate.
RankBrain takes it a step further by allowing the searcher to enter whole sentences into the search bar. In fact, the more words you use, the better the results. That’s RankBrain’s “thing.” It likes to read between the lines. It then holistically interprets the entire query, even deriving meaning from words that you might not think are important. These are called “stop words.” Words such as “about,” “almost,” “formerly,” and “except,” used in normal conversation, actually tell us a lot about the meaning being expressed by the speaker. RankBrain recognizes the importance of these words to understanding the greater meaning behind the request. Therefore, those words are no longer excluded from the search criteria. RankBrain processes them, then helps Google’s algorithms to figure out where to send them.
More about RankBrain