Understanding Semantic Search – Introduction for Beginners

I admit it.

SEO was so simple in the past.

All you had to do was include specific keywords in your content. Google would match them to someone’s search query, word by word, and BAM! You had a visitor.

One could almost guarantee results.

It’s so different now though.

Today you optimise your page for certain phrases only to see sites with different keywords outranking you.

Sigh.

But don’t get fooled by this. Google didn’t suddenly dropped search results quality, quite the opposite in fact.

The search engine got better, much better, at figuring out what users really search for.

Today, instead of just matching query to keywords, Google looks at other factors, such as search intent and context to deliver more relevant results.

This is called semantic search and you should learn as much as possible about it.

After all, it’s the future of search.

What is Semantic Search?

According to Wikipedia, semantic search aims to “improve search accuracy by understanding searcher intent and the contextual meaning of terms as they appear in the searchable data space, whether on the Web or within a closed system, to generate more relevant results.”

The aim for semantic search is to process information to discern search context and deliver a relevant answer.

Tony John at Techulator shares this example to illustrate semantic search:

Let’s say you’re sitting at home in front of a computer when someone asks you, “does it have windows?” It’s only natural to assume they inquire about your operating system.

But if they asked a realtor the very same question, there’s a big chance that they refer to a specific property listing.

And thus there would be a different intent behind the same question.

We humans can discern that intent based on context and give relevant answer.

Semantic search aims to emulate that behavior in a search engine.

Two terms are crucial in the above definition:

Intent. It denotes what the searcher is looking for.
Context, which describes everything that gives their search phrase a meaning.

Semantic search aims to identify both and deliver the most relevant results based on that knowledge.

So, How Does Google Do It?

Just like we humans look for and analyse available information to create context (a computer or realtor in our example above), so does a search engine. It considers a number of factors to establish both intent and context and deliver the most relevant search results.

Here are some of them:

A Searcher’s Location

A lot of queries are location dependent. And Google might use a searchers location to discern the context and deliver a search result.

Searching for a simple query like “What’s the weather today” will automatically return weather information for your location.

Search Intent

Google can also deduct search intent from the query itself.

For instance search for “best apple” returns listings related to the fruit.

“Cheapest apple” on the other hand lists Apple devices.

User’s Search History

Google can also use information a person has searched for in the past to discern their intent.

(Hey, how come Google knows I’m such a Machead?)

Spelling and other word variations

Synonyms, different spellings and other word variations also help Google to establish context?

In this Google+ post Mark Traphagen points how searches for “Google authorship gone from Google” bring results containing words like “dropped” instead of “gone”.

Concepts

Google might also look at the concept within search query. For instance, “traffic problems London” could naturally yield various results related to the topic:

List of closed off roads,
Accidents obstructing traffic,
Roadworks,
High traffic roads and many more.

Natural Language

The search engine will also look at the natural language, instead of processing a query as a string of related words.

Where to Learn More

I admit, the above just scratched the surface of semantic search.

There’s way more you need to learn before you’ll be able to avail of it’s potential in marketing.

Below is a roundup of resources that should help you learn more about how semantic search works and find out how you can unlock it’s full potential for your business.

General Information

In mid 2013 Simon Penson wrote on how Semantic Web changes everything for search. Even though his article relates to semantic web rather than search, it’s still worth reading to familiarise with concepts of semantics.

In this article, David Amerland answered most common questions regarding semantic search.

Seth Grimes expanded on those questions, presenting 11 different approaches to semantic search.

Just Briggs postulates that we should also stop working with pages and content and start working with what he calls Entities. In his words:

“An entity is anything, including real world objects, facts, and concepts, that has a number of documents associated with it. In the historical search models, a document is supported by other documents, but in the entity world, a conceptual object is supported by documents.

Keyword Research

Sujan Patel in this brilliant piece for SEJ shows how to conduct a three level semantic keywords research.

Copyblogger discusses how to create smarter content using semantic keyword research.

And Neil Patel wrote a tutorial on how to improve your rankings using semantic keywords research.

Content

Few weeks ago I published 5 ways to optimize content for semantic search.

Megan Totka also shares various ideas for optimizing for semantic search.

Conclusion

Search engines are changing. Instead of continuing to match query to keywords, they try to discern the searcher’s intent and context to deliver the most relevant results.

This is called semantic search.

And this is the future.

Now.