SEO used to be quite straightforward and somewhat technical: You needed to identify your target queries you wanted your site to rank for, create content including those keywords, and make sure that content is easily accessible by search crawlers.
Much of that process is still the same these days: We still start our optimization journey by identifying which keywords we want our content to rank for by evaluating demand and competition.
But it’s just the start…
Google is moving away from matching exact queries to documents in its index. It is now implementing semantic analysis and machine learning algorithms to evaluate queries in context and understand each individual searcher’s intent. Google’s algorithm can now provide solutions, not just list results matching queries.
Search engine optimization should be moving in the same direction: Instead of creating content matching identified queries, content needs to address related problems behind those queries. In other words, in order to remain effective, SEO should mimic what Google is already doing.
Here’s exactly what it means:
Use Semantic Analysis and Topic Modeling
To better understand what their users really want, Google uses advanced semantic analysis and topic modeling. Over the years, Google has analyzed tons of data and monitored billions of sessions to create a semantic model of the world giving them an idea of what lies behind any search.
In my article on semantic search, I am describing “Game of Thrones” topic example and how Google’s knowledge of it allows it to match a pretty generic query [ice fire] to a very specific idea, i.e. the books the famous TV drama is based on. Similar examples of semantic connectivity are all over Google’s search results pages:
Google doesn’t need a keyword-matching process to know that the TV drama has everything to do with:
- winter
- bastards
- dragons
- ice
- fire
[Google can guess what you are searching for without matching a single word from your initial query. Here’s one practical implementation of the new search algorithm: Finding what you need when you don’t even know what you are looking for]
This is where the concept of keywords is becoming somewhat obsolete. The SEO industry has entered a new age which we refer to as optimizing for “things, not strings” were we focus on topic modeling rather than exact-matching content to the strings of words.
Unlike Google, SEOs don’t have years of data to act upon though. But there are two ways to still optimize for things:
- Use Google itself: From Google Suggest to actual result pages that include “People Also Ask“, featured snippets, images, videos, etc. – Google gives us lots of clues as to what the actual answers their users are looking after. Make it a rule to scroll through search results Google generates and take note of the overall context and angles.
- Use semantic analysis tools. There are not too many options here but those that exist are pretty solid. Text Optimizer is the one I am using to understand any query concept. For any keyword you target, TextOptimizer will extract Google’s search snippets and apply semantic analysis to cluster it into related concepts and entities for you to include in your copy. You don’t have to include everything it suggests but the more concepts you manage to cover, the closer you get to Google’s model of the topic:
[Use Text Optimizer as a topic research tool to create a better-research more in-depth copy]
Recommended for You
Webcast, September 17th: 10X Growth & Relationships with Community Building
Embrace SERP Marketing
Apart from exact-match keyword optimization going away, search optimizers are facing another challenge: Google’s organic positions are losing value.
Google has slowly but surely moved away from minimalistic 10-blue links layout. The search giant is quickly turning into a destination, not just a discovery platform. It strives to give answers right away, often moving organic listings far below the above-the-fold part of the screen:
This means one important change for the SEO industry:
Goodbye, organic optimization. Welcome, SERP marketing
The SERP marketing concept means that you need to take each individual search result page as a whole and create a strategy around ranking in more search elements, not just 10 organic listings.
This creates a new, more diverse, and more integrated approach to SEO where you need to create and optimize all kinds of content assets, including:
- Images to rank in image carousels
- Videos to rank in related video boxes
- FAQ to rank in “People Also Ask” results and featured snippets, etc.
Our Featured Snippet Plus tool is the first-ever SERP marketing tool that helps you optimize for each query to dominate all search elements that show up:
This process includes implementing all kinds of structured markup to point Google to where you provide how-to content, recipes, news articles, etc. and generate more clickable search snippets.
Imply Advanced Engagement Analysis
Finally, user engagement (as well as positive user experience) is not just a conversion metric any more (which is important though).
It’s also your visibility metric as Google wants its users to be satisfied with the provided results. If your site doesn’t seem to satisfy Google’s users, you risk lower or lost rankings.
Conversion tracking will help you evaluate how good of a job you are doing meeting your users’ needs. Possible tools include heatmaps, usability testing platforms and advanced sales funnel monitoring.
Finteza is an independent web analytics tool you may want to give a try. It makes conversion monitoring easy and effective. You can easily filter your reports by a traffic source to estimate if search users seem to be satisfied with what you have to offer:
[Keep an eye on any negative trends here]
Conclusion
Google remains the most important source of online visibility. There’s simply no alternative to Google traffic. But the search giant is changing and it’s changing fast. In order to get found, businesses need to keep up with those changes by embracing new tools and new search optimization tactics.