02 Apr How Google’s Semantics are Changing Search
If Reddit is the front page of the internet, then Google is
the homepage. And it’s a role it takes very seriously. Since the infamous Panda
update of 2011, the search engine has made a mission of delivering the best
possible user-experience. Google wants to be able to provide the most relevant
and informative answers to your questions; answers that go beyond the standard
10 blue links, interpreting what your search queries mean and delivering
personal results. Helping it do this, and seemingly positioned at the vanguard of
search moving forward, is semantic search.
Search, but not like you knew it
“Semantic search works
on the principles of language semantics. Unlike typical search algorithms,
semantic search is based on the context, substance, intent and concept of the
searched phrase.” – Technopedia
If the thought of Artificial Intelligence (AI) makes you
uneasy then you might want to consider a different search engine, because this
is how Google is going to be managing your search in the future. Semantic
search is an AI engine designed to understand your search intentions and the
meaning of the query, rather than the words.
In the past, Google delivered results solely based on the
keywords entered – in other words it would take a stab in the dark. Semantic
search is much cleverer than this, recognising words with different meanings
and understanding the relationship between words placed together in a sentence
to interpret queries. Beyond this, the technology is also able to predict and refine
queries and extract entities as answers to deliver interactive results using
natural, conversational language.
Basically, by using semantics Google can read your mind…but
not in a creepy way.
Semantic search in action
You might never have heard of semantic search, but you’ll
have almost definitely seen it in action.
It comes into play from the moment you enter your first word
in the search field thanks to Google
Instant, which predicts potential results as you type and aims to save up
to five seconds per search.
It’s also evident in your search results. Let’s use “London cafés” as an example.
Using semantic search, Google assumes that you’re not
looking for a blog discussing the “history of the London café” but somewhere to
enjoy a warm beverage and a bite to eat and therefore serves up results of cafés in the area and articles related to London’s best cafés.
Google isn’t happy with just serving up documents; it wants
to delve into your mind and give you the most relevant information. But it doesn’t just stop at query prediction and location-based maps, no sir – it
wants arm you with knowledge.
The Knowledge Graph
Google’s Knowledge Graph represents the future of search.
The map and café listings are an example of it in action, but only a small part
of what it is capable of. Whenever you
perform a search, Google knows (or believes) you’re not just looking for a
website, you’ve come in search of answers. You want to learn, understand and
Whenever you search something, Google will deliver a mass of
Knowledge Graph-powered information.
Here’s an example of it working based on a search for “New
Notice that the Knowledge Graph (inside the red box)
delivers more information than you probably needed, including a description of
NYC pulled from Wikipedia, current weather, local time, population and even
places of education. From here you can expand your information further. For example, If I were to click on the image
of the Empire State Building in the “Points of Interest” section, I’m taken to
a new page of search results that features a carousel of buildings in New York,
as well in depth results on the building in question.
This is the power of the Knowledge Graph and it can go on
forever until you find yourself immersed in information on something completely
unrelated to your original search.
Here’s where I ended up in just three clicks from the “New
York” results page:
Go, Diego, Go! Anyone? Ask your kids.
If you’re intrigued by this Knowledge Graph, as well you
might be, here’s Google’s introductory video:
How does semantic search affect SEO?
By focussing on intent, semantic search isn’t just hammering
nails into the coffin of black hat SEO; it’s padlocking the thing too. Keywords
can be manipulated – people have been doing it successfully for years – but the
old data-influenced approach isn’t going to cut it any more. Words and phrases
have meanings and SEO will need to focus on them.
You can’t just think about a word, you’ve got to think about
how people will use it in a search. Take the word “salad”, for example. What are
people looking for when they search for this word?
- Types of salad
- How to make a salad?
- Ingredients for a potato salad
- Salad dressing recipes
The list, like the Knowledge Graph, could go on forever.
You’ve got to produce content that uses keywords in a natural way, based around
answering the questions of search users.
As a search user, the Knowledge Graph is an excellent
addition to results pages; however, it gives SEOs a problem. As if ranking for
keywords wasn’t already tough enough, you now have another competitor to
contend with in Google. The more information the search engine provides, the
less likely it is that a person will click through to a link, especially one
that isn’t a household name.
How do you overcome this issue and make Google a partner
rather than a rival? It’s pretty simple really, publish quality content.
Great content that informs and engages the user – combined
with highly-targeted, natural language keywords – is what Google wants, and when
Mountain View wants something it’s oblige or die.