Have you noticed how Google can answer almost any question you ask?
Let's look at the result of the following query:
Although we have not specified the name of “The Arc de Triomphe”, Google understood what we were talking about and showed what we wanted to know about it, its height.
This is possible thanks to “semantic search”. You may have already heard of this term. What is it? How is it affecting SEO? What changes?
Although semantic search sounds complex, it is actually not that complicated to understand.
In this blog we will explain everything you need, and you will be able to learn: what semantic search is, why it is important, and how to adapt your SEO for this type of search.
What is semantic search?
Semantic search is an information retrieval process used by modern search engines to return the most relevant search results. In other words, the purpose of semantic search is to go beyond the “static” dictionary meaning of a word or phrase to understand the intent of a searcher's query within a specific context.
The search engine learns from previous results and creates links between entities, so it can deduce the answer to a searcher's query, rather than providing several blue links that may or may not have the correct answer.
Why is it important?
Although there are countless variables at play, it's easy to understand the principles of semantic search.
Users do not use the same language as the desired content.
This happens, for example, when you heard an unknown song on the radio. Then you start searching for random lyrics of the song on Google until you find it.
Or the same thing happens when you talk to Siri, Alexa or the Google Assistant. What you speak to these assistants is often different from what you type into Google; keywords now become conversations.
There are many ways to express the same idea, and search engines must deal with all of them. They should be able to match the content of their index to your search query, based on the meaning of both.
Many searches are unintentionally ambiguous
In Spanish, as well as in many other languages, there are polysemous words, those that have two or more meanings. This is one of the main challenges of semantic search. This is arguably the most important challenge that semantic search is trying to solve.
The problem here is that words rarely have a definitive meaning without context. In addition to polysemous words, you can write countless nouns that can also be adjectives, verbs, or both. And we are still within the reach of literal meanings. It becomes even more interesting if we delve deeper into inferred meanings like sarcasm.
Context is very important in semantics.
The need to understand the lexical hierarchy and relationships between entities
If we perform the following search:
Google has to do several things to understand the query:
- Understand that Obi-Wan appeared in multiple films and series played by different actors
- Make connections with each other's wife
- Display search results in a way that reflects the ambiguity of “obi wan” but still answers the search question
As mentioned above, our queries often do not match the exact wording of the desired content. For example, “affordable” could be anything between cheap, mid-range, and reasonably priced, and knowing this is crucial.
The entities, in this example, are characters from movies and series (Obi-Wan), people with a specific job (actor), and people associated with them (partners/wives). In general, entities are objects or concepts that can be clearly identified.
The need to reflect personal interests and tendencies
If I perform the following query:
Google is going to show me the age of the current president of the country where I am, Mexico, and not those of other presidents from different parts of the world.
No matter how much we dislike the ways our personal data is used, at least it is useful to search engines. Google uses limited data, such as your location, along with your search history to provide more accurate and personalized search results.
But what's more fascinating is Google's ability to adjust search results based on dynamic changes in search intent.
If someone searches for “pizza” in their browser, they could be searching to order a pizza, how to make a pizza, or the meaning of the word “pizza.” But after years of serving search results to users and analyzing their interactions with those search results, Google knows that most people who search for [pizza] are interested in ordering pizza.
And as we can see, the first thing it shows in the results are different locations from which you could order a pizza.
How to adapt your SEO for this type of search?
There are several tips that should be followed to adapt SEO to semantic search:
- Target Topics
- Evaluate search intent
- Use semantic HTML
- Use schema markup
- Turn your brand into a Knowledge Graph entity
- Create relevance through links
Target topics, not keywords
In the old days of SEO, you might have ranked highly for several slightly different keywords, but all on the same topic. That is no longer the case. Google now understands that all of these searches mean the same thing, and ranks mostly the same pages for all of them.
You have to keep this in mind when creating content. The goal is no longer to rank just one keyword, but to cover a topic in depth so that Google will rank your page for many similar keywords.
However, you have to be careful. Addressing a particular topic doesn't mean you have to cover absolutely everything related to that topic or go into too much depth.
Evaluate search intent
It is possible to post content on a certain topic that does not align with search intent. This could even be content with great SEO, but that most people searching on your topic may not be of interest to you.
So before you start planning a new piece of content, it's good to look at the top-ranking pages on that topic to infer search intent.
Use semantic HTML
Let's compare the following HTML elements:
Semantic HTML adds meaning to code so machines can recognize navigation blocks, headers, footers, tables, or videos.
But semantic HTML is still quite limited. While it says, “this is an image, this is a footer,” it doesn't convey the meaning of the actual content. That's why we mark schemes.
Use schema markup
Schema markup is an additional way to mark up your pages. It is also known as “structured data.”
The vocabulary of Schema.org contains hundreds of types associated with properties. You can use them to mark up your content in a way that is easy for Google to understand without complex algorithms.
So instead of Google having to extract key information from something like this:
Horneará los muffins por 20 minutos. Estos muffins contienen alrededor de 305kcal por porción de 100g.
You could have it in a structured format similar to this:
tiempo de cocción: 20 minutos
calorías: 305
Which Google could more easily extract from.
So, when a user wants to know how long a muffin takes to cook, or how many calories it has, Google can give the information in the best way.
Build your brand into a Knowledge Graph entity
La Knowledge Graph Google is a knowledge base of entities and the relationships between them. This technology allowed the change from the relationship by keywords to the semantic relationship.
Turning your brand into a Knowledge Graph entity is a long-term consequence of building your brand and applying the rest of these tips.
Create relevance through links
Links have always been indicators of relevance.
Both internal and external links to relevant pages using natural anchor text help Google figure out what your content might be about, even before processing it.
Don't forget about semantic search when doing your SEO
Semantic search has been a major advancement in modern search engines. Now they are capable of answering almost any question asked with relevant and valuable results for the user. Semantic search is possible thanks to very complex technologies and algorithms, but it is not that difficult to understand how it works and what its principles are. Nowadays it is a requirement to consider semantic search when perform SEO, don't forget to take it into account.
