Ten years ago, SEO strategists around the world followed a relatively similar process.
Step one, do keyword research. Step two, randomly write these keywords about five billion times in the text on a page. And step three – rank number one for that keyword.
I hate to interrupt you, but it is no longer the case.
Several algorithm updates such as Hummingbird and RankBrain have spawned a new concept: semantic search.
While this may remove jobs for black hat keyword stufferers, SEOs who prioritize the importance of a good customer experience can take a sigh of relief that Google is now on their side.
Google and other search engines are constantly striving to satisfy the searcher with the most accurate results – and this is where semantic search comes in. In other words, it combines search intent with the context of your content to deliver the most relevant and helpful results.
How does this affect search traffic with these updates? And what do SEOs have to consider when developing further?
That’s what I’m going to cover in this article.
What is semantic search?
First, let’s dig deeper into how semantic search works.
Semantic search is the process that search engines use to understand the intent and contextual meaning of your search query in order to provide you with results that match what you want.
In other words, the semantic search is aimed at knowing Why You’re looking for those particular keywords and what you want to do with the information received.
Important note: You shouldn’t confuse semantic search with Latent Semantic Indexing (LSI), or what some call semantically related keywords. LSIs can help provide context about your content (which thus helps match search intent), but semantic search is so much more than that.
If we take a holistic view of semantic search, here are the factors that determine how it works:
1. A user’s search intent.
The term “search intent” refers to the reason you are doing a search query (or, to put it in layman terms, why you google something). Most often, you want to buy, find, or learn something.
For example, when I search for “content marketing”, Google returns results around the definition of content marketing because the intent is quite broad:
However, if I search for “how do I get started with content marketing” instead, Google will Not Provide definitions of content marketing because my intent is different:
Take away: For all content marketers and SEOs, the big lesson is that you need to keep search intent heavily in mind when choosing keywords and creating content. Even if you have content that is ranking well, if it doesn’t match the search intent, the user will leave the page – and that certainly won’t help with conversions.
2. The semantic meaning of search terms.
The “semantic search” was coined based on semantics or the study of the meaning of words and phrases in specific contexts and the relationship between those words. When it comes to search, semantics refer to the connection between a search query, associated words, and the content on a website’s pages.
All of these factors together help search engines understand what the search queries mean beyond a literal translation so that they can display contextual results.
For example, if you were looking for “wedding dress,” the associated words might include “wedding,” “cake,” “bride”, and “dream”. When searching for “clothes” the related words can be “beautiful”, “knee length” and so on.
Take away: When choosing the keywords for your content, I recommend creating so-called “keyword clusters” or groups of related keywords. These clusters are directly related to semantic search as they ensure that your content covers a wider range of the topic. And with a greater reach, there are several keyword rankings per page.
Other factors related to semantic search
Although the above two are the main factors, these factors also affect semantic search:
- Selected snippets: Featured snippets are based on giving the searcher the most direct and helpful answer.
- Rich results: These also affect semantic search through content such as images, and you will see how to do this in the example in the next section.
- Voice Search: Voice search queries are usually very straightforward and involve natural language, longer phrases, and question words that help how search engines process results.
- RankBrain: Based on machine learning technology, the RankBrain algorithm helps Google understand the sentence of the first instance that satisfies the query and its related concepts, phrases and synonyms.
- Humming-bird: The focus of the Hummingbird algorithm update was to provide better results for voice search, conversational language, and search for specific people.
Examples of semantic search
To give you a clear idea of how semantic search works, here are some specific examples.
Here I searched for “order pizza”, so the results tend to be local searches:
Here I googled “Make a pizza” and see rich results with recipes:
Just googling “pizza” will likely still get local search results because more people are searching for it assignment instead of making them yourself. However, if my search history is filled with pizza recipes, my results for “Pizza” will likely be recipes too because of the personalization component.
The semantic search basically affects all results that a user receives. A website will only be delivered as a result for a specific keyword if the content on the side coincides with that context this search query. Results for “Make Pizza” include ingredients, preparation time, and so on, while “Order Pizza” includes locations, delivery, and prices.
Interestingly, breaking news also affects search results. Before the pandemic, a search for “Corona” would usually have returned the brand of beer, but after the spread of COVID-19 you mainly get results on the virus.
Another example is Jeff Bezos. If you search for his name, you’ll get a knowledge graph, general information, and breaking news underneath. However, if something big happened to Jeff Bezos recently, see the top stories first.
How Google uses semantic search
The bottom line of Google is to give users the best search experience possible. To do this, they use semantic search to:
- Identify and disqualify inferior content.
- Gain a better understanding of user search intent. Example: Is the user looking for a specific page? Or would you like to do more research on a topic?
- Formulate answers to questions.
- Determine which relevant data should be drawn from the Semantic Web
- Understand websites and pages in terms of topics rather than keywords.
- Integrate Google technologies in which semantic search plays a role, such as Knowledge Graph, Hummingbird, RankBrain, BERT.
- Format the data appropriately for inclusion in search results.
- Connect with queries of all sorts of meanings when the search intent is not clear.
This is how you use the power of semantic search to your advantage
Simply put, if your content doesn’t have a semantic relationship to the search query, it won’t appear in search results. The simple solution to this is to adapt your content to the search term in combination with the right strategy.
To be on the right side of SEO when it comes to semantic search, I recommend that you do the following:
- Focus on topics, not keywords.
- Make sure you understand user search intent: is it to buy? to reach a certain side of a brand? Learn?
- Build relevance through links (both internal and external).
Use schema markup.
- Use semantic HTML like
- Answer all relevant questions about your topic.
- Be answer-based and structure your sentences so that they are easy to understand.
When you cross these off your list, you get a powerful one-stop SEO strategy with semantic search support.