There was a time when web copy written primarily for users or for search engines was quite a bit different. Skilled copywriters could get their point across to both in an unobtrusive fashion. The less skilled stood out like sore thumbs.
As Google's approach to search and spam control evolved, a great many things changed. Meta keywords were deprecated, the notion that it was necessary to embed keywords repeatedly slowly faded (a comment here and there about keyword stuffing may have helped). And of course, Google spokespeople repeatedly voiced terms like quality and naturalin an effort to break us of the habit of wooing their algorithms instead of our users.
What's Different Now?
Glad you asked! Quite a bit has changed, actually. At least in the most effective way to write your copy.
Say your page is about "jaguars". There was a time when you'd absolutely have had to use "jaguars" in your content a time or two. And at that, if someone was searching for Florida football teams or Guatemalan predators, your page might have an equal chance of appearing in either SERP.
Of course, more specific terms like "Jacksonville Jaguars" or "Central American jaguar" could eliminate some confusion, but "jaguar" was still the key term. Why? Because that's what the search algorithms needed, in order to be able to connect the dots.
Now, though, the algos have progressed well beyond simply synonyms. They're beginning to be able to grasp the concept behind a piece of content and categorize it. While they've not yet arrived at a high degree of semantic capability, they've come a long way, baby.
With "jaguar" being a term easily connected to a football team, an automobile, an operating system or a jungle cat, the algorithm might understandably err on occasion. But with other related terms embedded in the same content, such as "running back", "XK", or "carnivore", today's search algos are able to zero in on a target with ever-increasing accuracy.
Even if "jaguar" appears nowhere on the page.
What About Keywords?
Relax, I'm not saying keywords are obsolete. I am saying, however, that they're less crucial than they once were. In fact, I've even been able to rank pages for terms which didn't appear anywhere on the page. by simply focusing on a clear concept.
Keywords still matter, for a couple of reasons:
- Search algos are still babies, barely setting out on their learning journey. So using "jaguar", "predator", "jungle cat", "carnivore", and "claws" in the same content will help the algos learn the relationships.
- Users will still enter keywords as search queries, and at least for the foreseeable future, relevancy between the exact search query and the target content is likely to remain an important factor – but not necessarily essential.
What Will This Mean?
OK, here's where the controversy may kick in, because not all SEO professionals are in agreement. In fact, there are even some search engineers that still argue that semantic search is a long-term goal. In my opinion, though, it's not long term at all, perhaps just a handful of years away. And from what I'm seeing, we've already got a good start.
The beauty of this is that it will gradually eliminate a major problem in content generation. As this semantic capability continues to develop, it will no long be necessary to write content for both user and search engines. Copywriters will be able to concentrate on the user, and when presenting the content in the best possible light, the content will be as effectively presented to the search engines as it is to the users.
As you can imagine, removing the temptation to repeat keywords numerous times should help eliminate some of the crappier content found on every corner. And as the algos continue to learn, it will enable copywriters to produce higher quality, user-centric content that reads more naturally and informs better.
It will mean that content will be judged by its predominant concept... as a whole, what is the piece about? Even to an algorithm, mention of a "sleek nocturnal predator eyeing its prey from the edge of the jungle" will obviously not refer to an exotic automobile or a running back.
When an entire document speaks to the characteristics of a jaguar, the algorithms will be able to learn relationships of those characteristics to that cat, and put them to future use. That learning process will gradually help interpret not only what individual documents are about, but when those learned relationships are fed to the new Hummingbird algo, it will enable more effective interpretation of search queries.
More relevant SERPs. Because regardless of how you may view Google's intentions and motivations, SERP quality is what best supports the company's primary focus: selling ads. High quality SERPs = more users = more ad sales. There's no conflict between the search giant making money and delivering relevant search results – on the contrary, the two are inextricably bound.
Twitter Canada MD Kirstine Stewart to Keynote Toronto
ClickZ Live Toronto (May 14-16) is a new event addressing the rapidly changing landscape that digital marketers face. The agenda focuses on customer engagement and attaining maximum ROI through online marketing efforts across paid, owned & earned media. Register now and save!