Semantic Search: What is it and Why is it Important?
By Sarah McInerney | July 8th 2020
By Sarah McInerney In this post we are taking a look at the subject of semantic search. What it is, why it’s an essential part of search engine optimisation campaigns and how you can use it to enhance your visibility in the search engines.
Semantics is something we looked at a while back when we explored the topic of related keywords and how they impact upon SEO content.
Semantically related keywords enrich online content and help to show the search engines that web pages are relevant to what the user is looking for. In other words, they help put things in context so as to avoid confusion over the subject matter. Handy when you have topics that would by the title appear to be the same, but in reality couldn’t be more diverse. We’ll go into that in more detail later. First, let’s start with the basics. What is semantic search?
What is semantic search?
You know we talk a lot about search intent and how important it is. It’s a vital tool in the mission to rank higher in SERPs. When someone types something into the search bar, are they looking to buy something? Or do they want to learn? This is at the core of search intent. And this is where semantics really matter.
Semantic search involves search engines analysing the intent and in-context meaning of search phrases ahead of delivering the most relevant results. Google for example uses BERT – Bidirectional Encoder Representations from Transformers – to better understand search queries.
Here’s a summary of what search engines do when they attempt to understand natural language in the same way we humans all inherently do:
- Understand search intent
- Cross reference the context
- Analyse relationships between words
When you’re chatting with a friend, and they ask you a series of questions, you’ll instinctively know how they relate to each other. For example, if you asked them, ‘what was the largest pizza ever made?’, and then they asked you, ‘where was it made?’, you’ll know they are asking where that pizza was made.
Up until a few years back, a search engine would never have been able to see the connection between the two questions, so it would have simply returned results for ‘where was it made?’ with no reference at all to the world’s biggest pizza. Keywords were what search was all about. But now things are different, because we have semantic search.
Semantic search makes it possible for search engines to distinguish between different things, such as people and places. Now Google can interpret search intent by looking at factors like user location, search history and spelling variations.
For example, if you were recently researching wildflowers online, and then searched for ‘primrose’, Google would assume that you most likely wanted to find out about the wild primrose plant, rather than the garden retailer or the jewellery brand. Very clever, and totally focused on delivering a better user experience.
The history of semantic search
The history of search helps us understand semantics a little better.
‘Things not strings’ was Google’s catchphrase back in 2012 when it introduced its Knowledge Graph. This huge database of public information was the search engine’s first foray into placing the importance of entities and context over keyword strings.
In 2013, the Hummingbird update took things a step further by making sure that pages better matched search intent and context. And the pages that hit that target were rewarded with boosted rankings.
By 2015, RankBrain, Google’s machine learning tool, was using artificial intelligence to understand user intent at a deeper level. Because it’s is a system that’s continually learning, RankBrain keeps on top of the best performing search results, working out the parallels between those pages that users find the most useful. And it’s improving all the time.
Why is semantic search important for SEO?
SEO experts know that because of the rising importance of semantic search, and the growing intelligence of Google and the other search engines around analysing user intent and search context, it’s vital to factor semantic search in to online strategies.
So, how to use semantic search to your best advantage to enhance visibility? Time for some practical content marketing advice.
Go conversational
Almost 50 per cent of web searches are voice searches. Speaking your search query rather than typing it uses a much more informal and natural tone, plus you tend to get to the point faster.
So, if you’re optimising your content for voice search, which is highly recommended given that half of searches are done this way, you’ll need to factor in a concise question right at the top of the page. One that responds clearly to a common query. Once you’ve done that, this is where semantics come in.
What you need to do next is include related words and topics in the rest of your content. These will help the search engines understand the context of your content so that they know it’s relevant to the search. This is semantic search.
Here’s an example…
You’re selling European cruises. You create a piece of content that addresses the common query, ‘What are the most popular cruise destinations in Europe?’. You quickly answer the question right at the start with a list, perhaps including back-up statistics illustrating how many people took cruises to those destinations in the last year so that readers trust your information. Later in the piece, you include related content, such as things to take with you on a cruise, how to dress for a European cruise, how much does a European cruise cost, that kind of thing (see below for where to get ideas). So it becomes the ultimate guide with a host of related information, not just on cruise destinations but on everything to do with the European cruise.
Switch to topic research
If you’re still creating content around keywords, it’s time to knock it on the head. Replace this approach with topic research and any SEO expert will tell you that you can really start taking advantage of semantic search.
Make your objective to generate unique, all-encompassing, top quality content. Content that addresses the real questions people are asking. That satisfies your audience’s thirst for in-depth information and thought leadership.
Wondering how to uncover those real questions?
Here are some research tools to help you:
Answer the Public – a tool that brings Google’s auto suggest to life, not only showing you topics, but also some of the intent behind them. Genius!
BuzzSumo Question Analyzer – discover the most popular questions on any topic courtesy of a database of real questions from platforms like Amazon, Quora, Reddit and other forums.
Ahrefs Content Explorer – helps you find the most popular content around any subjects based on organic traffic, social shares and backlinks.
Make search intent a priority
If you know the search queries that led people to your website, it will make it easier to come up with the topics that will work best for content creation.
Google Analytics or Google Search Console can both help you here.
Make a list of the keywords and then split them according to user intent. Let’s take the European cruise scenario as an example.
Queries such as ‘European cruises from Southampton’ would indicate intent to discover what destinations can be reached from the port of Southampton. Whereas ‘European cruise deals’ would suggest an intent to book a cruise.
Semantic search: putting people first
Semantic search must involve putting people at the heart of your content marketing campaign, a crucial element of any SEO strategy. Focusing on what people truly want to know is essential, as is creating relevant, high-quality material. Technical optimisation is equally crucial of course. It’s a fine blend that the team here at London SEO consultant Figment has honed to perfection.
If you are looking to work with a SEO agency London businesses are happy to rate and put their trust in, look to Figment. To discover how we can help you grow your business by enhancing your online visibility, you are welcome to get in touch.