How to optimise your keyword strategy for semantic search
One of the essentials of content marketing is that performance on search engines is critical to a business’ success online. Search engine optimisation (SEO) is therefore a critical component of any digital strategy. Without it, the best content can simply go unread.
Search optimisation and Google
While there are many search engines out there, the undisputed king of the hill is Google, which processes some 40,000 search queries a second, according to InternetLiveStats. This is an absolutely enormous share of online traffic, amounting to over a trillion searches every year.
Even so, performance on Google is the subject of much mystery, with the company’s engineers constantly refining and polishing the platform’s algorithms in order to provide the best possible user experience.
An enormous part of this is ‘semantic search,’ which focuses on user intent, specifically by considering language in the way we mere mortals use it. But what does semantic search really mean, and how can content marketers use its rules in order to get websites to the front page?
Google processes roughly 40,000 queries every second, and over a trillion searches each year.
What is semantic search?
Google’s own dictionary defines semantics as “the branch of linguistics and logic concerned with meaning,” which goes a long way towards explaining why it’s so important to search and SEO.
Semantics deals not just with language, but what we take individual words to mean and how we string them together as requests for information. The focus is therefore less on the individual words or phrases used, and more on the search intent behind them. After all, language varies wildly based on the specific situations we find ourselves in, with word choice on a search engine likely to be very different to other areas of life. Semantic search aims to dig through this surface layer of language in order to better understand the intent beneath.
A great example of semantic search comes from Moz, which uses a simple query like ‘Jennifer Lawrence,’ which has been searched over 7.5 million times, to illustrate how Google uses data.
As Moz explains: “Google provides news, photos, facts, social media accounts, and movies all related to Jennifer Lawrence.” This ensures the broadest possible range of queries are answered, despite the search itself offering no real indication of what the user is looking for.
Google’s Knowledge Graph
The pinnacle of Google as a semantic search engine is its Knowledge Graph, If you’re not familiar, this is the information box that pops up at the top of certain search results. For example, if you search for ‘weather in Sydney,’ the Knowledge Graph will display the temperature and forecast, as well as key information such as precipitation and wind. Similarly, if you were to search for something as simple as ‘NSW cities,’ the information displayed offers a list of cities, as well as answers to common questions such as ‘is Sydney a coastal city?’ and ‘Where is the best place to live in New South Wales?’
What the Knowledge Graph is doing is using the principles of semantic search to provide information that it feels best meets user intent. There’s nothing in the language of ‘NSW cities’ to suggest interest in the best places to live within the state. However, based on previous searches, common trends and a host of other factors, Google’s algorithm is able to guess that it’s a likely motive for the query.
The components of semantic search
From a purely semantic perspective, the most important information that Google uses to provide results that meet user intent is contextual data. This includes everything from where in the world a search is being made through to an individual user’s search history. This information is combined with the traditional ranking factors that affect a position on Google (such as the quality of content and use of specific target keywords) to create a tailored set of information.
In addition, semantic search accounts for human error. We’ve all felt the shame of spelling something wrong in a Google search and being faced with a taunting ‘Did you mean?’ followed by the correct word. In many cases, Google won’t even bother with that step, and will simply show results for the correct search term, with the option to revise the search to its original gibberish if you so desire.
All of these components are designed to make searching something on Google as simple and intuitive as talking to a (very smart) friend.
Semantic search is designed to make using Google as simple and intuitive as talking to a (very smart) friend.
Semantic search and SEO
In the past, search engine optimisation was a relatively simple process. Performing well on Google relied on simple strategies such as backlinking and including as many keywords as possible. To achieve this, content marketers would use a keyword planner to identify the specific words or phrases that a search engine was looking for, and include this wherever possible.
In many cases, this led to the practice known as ‘keyword stuffing,’ where the quality of content writing was set to one side in favour of huge volumes of specific keyword phrases. This resulted in a poor user experience, making it difficult to find relevant information on the front page of Google.
As a result, the engineers behind Google moved away from a pure focus on the presence of keywords, instead favouring context and intent when compiling results.
As the Search Engine Journal puts it: “You need to understand what keywords mean, provide rich information that contextualizes those keywords, and firmly understand user intent.”
This underlines the fact that keywords are still a vital part of SEO. What’s changed is that rather than simply being phrases that relate to a particular topic, keyword optimisation now focuses on what people are searching when they want specific information. These queries are known as ‘semantic keywords.’
Understanding semantic keywords
Semantic keywords are used to provide a clearer indication of what information a particular piece of content delivers, matching up with a specific search. They usually include a traditional or core keyword, but add context and intent.
For example, if a core keyword is ‘content marketing,’ a semantic keyword could be ‘what is content marketing,’ ‘how to learn content marketing,’ or ‘content marketing experts in Australia.’
As queries, each of these phrases is looking for different information, which wouldn’t necessarily be delivered by a Google search for simply ‘content marketing.’
As CopyBlogger explains, semantic keywords are: “Building a larger picture behind [a core term] They’re getting to the real question the searcher is trying to answer.”
For businesses, the takeaway here is that including well-placed semantic keywords in content can show Google’s algorithm that a particular website meets a user’s intent when they search for that particular phrase.
Identifying semantic keywords
Of course, using the right semantic keywords for SEO isn’t as simple as placing a few extra words on either side of key terms. Instead, semantic keywords need to relate to what real users are searching for. In some cases, these searches may not even contain the core keyword you’d expect.
An example of this is a user looking for information about home air conditioning. While you’d expect the core keyword to be ‘air conditioning’ and semantic searches to include phrases like ‘best air conditioning unit,’ there are all sorts of different ways a user might structure their request to Google. They could search for something as simple as ‘how to keep home cool in summer,’ but if that semantic keyword phrase isn’t included in an air conditioning manufacturers’ website, they likely won’t show up in the results. This is known as a thematically related keyword, and every content marketer should make sure to include some when creating an SEO strategy.
The best way to identify semantic keywords is deceptively simple – think like a human. If you were looking for a certain website or piece of information, what would you search in Google? With this in mind, it’s far easier to align with user intent and impress Google’s algorithm.
Even so, it isn’t always easy to compile keyword clouds that account for every possible variation on a search. Fortunately, there are all sorts of different tools available that content marketers can use during their keyword research.
Many of the top tools are made by Google itself, with Google Trends arguably the best place for understanding what people are searching and why. The platform provides a wide range of data that includes how popular certain searches are, as well as related queries. Of course, to see how semantic keywords work in practice, you can also simply look at a Google search, where you’ll see a list of related, semantic searches.
In order to be successful on Google, content marketers and the businesses they work with need to understand the value of these semantic searches, as well as how to turn them into semantic keywords that fit naturally into content. As Google’s algorithm continues to become more advanced, failure to do so could result in slipping further and further away from the invaluable real estate of the front page results.