One thing has remained a constant in the search industry: frequent – and often dramatic – changes in technology.
As SEO professionals, it’s necessary for us to continue to evolve in order to remain effective.
Today, we’re at a point where this is occurring faster and more dramatically than ever before.
While some of the previous changes to search affected specific tactics, the ones we will face in the future — changes that we already face on a smaller scale now — will tend to impact search on a more strategic level.
These technological advances will completely revolutionize search.
There’s a Revolution Calling
Some of you may be relatively new to search and haven’t yet seen a groundbreaking change in the industry.
A much smaller group of industry veterans likely remember what search engine optimization was like in the days before Google, when links weren’t even a ranking factor.
But the dramatic change that we’ve seen over the past two decades will be just a drop in the bucket compared to the change we will face in the coming years.
Technological advances will completely revolutionize the industry, creating a tremendous opportunity for proactive and intelligent SEO practitioners, but likely also create chaos and collateral damage on a scale that the industry has not yet seen.
If we expect to remain competitive in the coming years, it’s essential that we’re aware of these technologies and that we understand them as best as possible. That will require a combination of:
- Analyzing relevant patents or closely following the work of those who do, like Bill Slawski and Roger Montti.
- Consistently conducting large-scale experiments to prove or disprove theories.
That last part is critical because while a patent may say that something will work a certain way, the results in the real world are often very different.
That being said, these are the technologies that will revolutionize the way we think about search — and probably far sooner than we expect.
1. Artificial Intelligence & Machine Learning
Talk of artificial intelligence and machine learning have hit a fever pitch in the last few years as the technology behind them have advanced at a rapid pace. However, as far as they’ve come, they still have a long way to go.
I like to compare the behavior of AI to that of a drunk toddler. Sometimes it gets things right, but just as often, it gets things wrong, and every so often, it does something completely embarrassing, ridiculous, and even terrifying.
Artificial intelligence, powered by machine learning, has the potential to be an amazing and positive force in powering search engines, but it still has a lot of growing up to do.
Right now, in its infancy, it’s still crawling around in Garanimals PJs with a pacifier in its mouth. In a few years, it will be more like an angsty teenager screaming “You don’t understand me!” while eating Tide Pods — but it will also be a lot closer to delivering useful results more consistently.
And by results, I don’t just mean spitting out 10 choices, or even one, in response to a query plugged into a search field. I mean that it will go so far outside of the traditional concept of a search engine by anticipating and predicting our needs long before we even think about them.
I’ll call this predictive search.
Here’s an example: I can sometimes be geographically challenged, so when I leave an appointment, I typically open my map app and click my office address to guide me back.
This is a perfect example where AI and machine learning can shine.
When my appointment is over, rather than me having to manually open my map app and enter the address of my next appointment, Google can evaluate whether I’ve been to the address of my next appointment before, and spring into action by either:
- Automatically opening my map app with the next address on my calendar ready to go if it’s somewhere new,
- Offering a prompt if it’s somewhere I’ve been a few times,
- Or simply not opening at all if it’s somewhere I go frequently.
But this technology can go far beyond that, offering assistance in more creative ways, and through channels most people have never even considered.
Sure, we’re all familiar with conducting a search through a website, app, and even devices like Google Home, but what about when search engines start to use data from a variety of sources to predict our needs before we ever conduct a search?
An Exciting (Or Terrifying) Future?
Imagine if your children came home from school sick — if you’re a parent, this doesn’t require much imagination because as much as we love them, we have to admit that kids are like disgusting little germ factories, and the schools we send them to for a large portion of their day are basically turbocharged Petri dishes filled with near weapons-grade pathogens.
So, predictably, you succumb to whatever mutated plague they happened to bring home from their fellow germ factories, and you end up calling in sick the next morning.
Shortly after emailing your boss to tell him that you won’t be in, the clients and coworkers you were scheduled to interact with later that day all receive an email too — but not from you. AI identified and interpreted your email message to your boss, and automatically informed everyone else on your calendar for that day that you wouldn’t be in.
Later, after a brief NyQuil-induced coma, you emerge from your bedroom, still bundled in pajamas and a robe, with hair a mess and breath that could gag a opossum, to answer a knock at your from door.
There, on the ground before you, sits a small package from Amazon that you don’t recall ordering. Tearing into it, you find three bottles of kombucha tea, a package of throat lozenges, and a bottle of echinacea and vitamin C.
“Where did this come from,” you wonder? Then you remember that you discussed being sick with your wife last night in the same room as your Alexa. (Mental note: move Alexa out of the bedroom.)
As creepy as this scenario may sound, it’s all well within the realm of possibility for artificial intelligence, and I believe we’re not too far off from it becoming reality. A reasonably conservative estimate could put it as soon as five years. However, Jim Hedger and I discussed this topic on a recent episode of Webcology, and he believes it could be closer to a year or two.
The hardware is already here, and thanks to the rapid adoption of personal assistant devices in recent years, such as Google Home, Siri, and Alexa, comfort with the idea of a device that’s always listening is becoming the norm.
The only thing missing is AI that’s smart enough to perform these type of tasks well on a consistent basis.
That’s advancing at a shocking pace thanks to the existence of appropriate hardware, widespread adoption of the technology, and the use of machine learning, taught through voice search, to train or teach the complex artificial intelligence algorithms how to perform.
2. Voice Search
I’ve been in this industry long enough to remember when it made sense to create an individual page for every possible variation of a keyword, including common misspellings.
As you might imagine (or as you may remember), this created a lot of garbage online.
Fortunately, the thinking on keyword research has evolved dramatically over the last two decades, first slowly transitioning from a rigid keyword-based approach to a more natural topic-based approach, then blending that topic-based approach with natural language. By natural language, I mean search terms that model how you might speak to another human being.
This most recent evolution is largely thanks to voice search.
Voice search has technically been around for a long time, but it didn’t really take off until about 2013 when smartphones like the iPhone and Android became common. That put the power of the internet into the hands of the average person no matter where they happened to be, but without the convenience of a keyboard and mouse, making voice search an easier and more convenient option.
While it may seem like one type of search query is the same as any other, voice search is an entirely different creature because the results are so intent-driven.
For example, if I search for a restaurant we have here in Tampa called Texas de Brazil from my desktop, a search engine knows that I’m probably looking for information about the restaurant, so the first organic result will be their website. If, on the other hand, if I conduct that same search using voice search via a mobile device, I probably just want to know how to get there.
Ultimately, you need to think beyond the search query and think about what problem a user is most likely trying to solve with that search query. Provide that and your likelihood of performing well in voice search will increase.
As voice search continues to grow, it will become an increasingly valuable component in the machine learning that enables search engines to refine the “brains” behind their AI.
Natural language queries combined with a growing understanding of related entities will create an AI that is capable of thinking rather than simply retrieving data that matches a keyword-based query.
Unlike keyword-based queries, natural language queries, in which we’re speaking to a search engine through a device, give search engines an opportunity to learn from users and respond in real-time by analyzing a variety of factors, which could include:
- Does the user appear to have found what they needed from the result provided, or did they quickly return to make another query?
- If the user didn’t find what they needed from the result provided, did they then ask for the next result, modify their initial query, or did they rephrase the query entirely?
- Does the user switch from voice to traditional search? Did they appear to find what they needed at that point?
- Is the user’s voice relaxed and at a normal conversational volume, or agitated and raised? Has it changed at some point during this process?
Machine learning can determine which results best satisfy users under which circumstances and then pass that data back to a centralized repository where it can be compared with data from other devices; data that shows a statistical improvement in delivering what the users want can be implemented into the core algorithm in real time, resulting in an algorithm that teaches itself at an exponential scale.
But the ability of voice search to teach artificial intelligence how to understand our language goes far beyond simply returning relevant results
This technology could also be used to “crawl” what I would imagine has to be trillions of terabytes of data available from podcasts and video, which is currently completely invisible to search engines.
It would be a game changer because instead of relying on titles, descriptions, and tags, the actual audio content within MP3 and video file could be analyzed, opening up a wealth of new content to searchers.
To take this to the next logical step, it could also evaluate the individual frames within a video file in the context of the audio content of that video, and begin working to understand video as well.
This would enable whichever search engine first developed this technology to provide a wealth of information that their competitors couldn’t, and would provide additional leverage in search for proactive marketers who have invested time into building a library of useful podcast episodes.
3. Mobile Search
As the foundational technology that enabled voice search to become useful, mobile search has already revolutionized search, but it will continue to do so, perhaps on an even larger scale than it has already.
The most obvious role it has played is in enabling users to more effectively navigate the web in an efficient environment. No more pinching and zooming — instead, it gave us responsive websites that can dynamically adjust for optimal use on whichever device it’s being viewed on.
But mobile search goes beyond smartphones and tablets. Think of small wearable devices, like smart watches, all the way up to internet-enabled vehicles. Basically, any device that could enable users to conduct a search on the go.
We’re already at a point where over half of all searches are conducted on a mobile device, and that percentage will continue to increase as we adopt new types of devices in greater numbers. As that happens, mobile search becomes increasingly important.
So what matters in mobile search today?
Search intent plays a huge role, both because there is limited screen real estate on which to display results, and because when a user conducts a search on a mobile device, there is usually a very specific and immediate need.
Unlike on a desktop, where a user may be simply gathering information, a user conducting a search on a mobile device is likely looking for, or en route to, a particular location.
For example, if I search for Thai restaurants from a mobile device, it’s a safe assumption that I’m looking for one near my current location, and if I search for a particular Thai restaurant from a mobile device, I most likely want reviews and directions.
Speed is also a critical factor for search in general, but it’s especially important in mobile search because mobile devices lack the bandwidth and processing power of desktop computers.
Frankly, this is an area that most people fail miserably in because a lot of web designers and developers lack the expertise to properly optimize a website for speed.
This means optimizing media files, reducing HTTP calls, and leveraging caching and Content Delivery Networks (CDNs).
But that’s just barely scratching the surface.
Truly optimizing a website for mobile is equal parts art and science, and often requires a fair amount of trial and error to get it just right.
If you want to do a deep dive into the future of mobile search, Cindy Krum recently published a comprehensive four-part series titled “Mobile-First Indexing or a Whole New Google?”
But mobile search soon will go well beyond traditional thinking, and even beyond bleeding edge thinking of the day.
While today we may think about websites when we discuss mobile search, in the near future, they may not even be a factor.
Think this sounds ludicrous?
Remember, it was just a few years ago when most people thought catering to mobile was insane. Now mobile accounts for more than half of all web traffic.
How Could Websites Become Irrelevant to Mobile Search?
Current technology requires search engines to crawl, parse, and interpret every individual webpage.
This is time consuming, costly, and inefficient.
While schema markup is helping to make this more efficient, it’s still not an ideal situation.
What would be ideal, at least for search engines?
4. Content Hosted by Other Entities
If we hosted our content on servers provided by search engines, the need for them to expend massive resources crawling the web disappears because they would already have it.
Before you discount this idea as crazy or dangerous, you need to realize that it’s already taking place.
While this clearly benefits search engines, it also provides tremendous benefits to users and marketers as well.
The question is will it be used for good, or will the search engines abuse our content, cutting us out for their own gain?
I think the risk is high enough, especially considering Google’s past history, to scare enough people away from the idea entirely.
For now, at least.
What Does the Future Hold?
In the same way that many video rental stores ended up with an inventory of useless Betamax players and video cassette tapes, it’s inevitable that you’ll invest time, money, and resources into search technology that eventually gets dropped or changed dramatically.
Those of you who are old enough to understand that reference have also probably been around the SEO industry long enough to have already seen examples of this, such as Google authorship.
The technology that I’ve discussed in this article is already here. Some aspects of it may be at the bleeding edge, but it’s going to be advanced and in common use long before most people have a chance to react.
The key to success in the SEO industry over the coming years will be to adapt to this new technology early and aggressively.
In some cases, you’ll find yourself at a dead end, but that’s the nature of early adoption to any technology.
More SEO Resources:
Featured image created by author, June 2018