Take a moment to think about how smart Google’s search engine is today. In the not-so-distant past, you’d type in a keyword and your search engine of choice would simply look for pages containing that particular keyword.

Nowadays in 2016, Google actively tries to figure out what it is you’re looking for. That means being able to:

• interpret the searcher’s context (including previous searches, location, etc.)
• interpret the wider context of the query (for example, current news events)
• recognise the synonyms that match up with the search phrase
• spot and correct any spelling mistakes the user has made, and
• review the quality and relevancy of the content on each of the billions of webpages on the internet.

Now remember that Google delivers your results in less than a second. That’s a powerful algorithm – and as you might expect, Google is fairly tight-lipped on how it all works to deliver great results time after time. However, over the years, it has revealed some fascinating insights into the technology that powers its search engine.

Let’s look at a few of the clever features behind Google’s search engine algorithm, and how you can best leverage them for your own SEO performance.

RankBrain

Google dropped a bomb in October 2015 when it not only introduced the world to the existence of its RankBrain technology, but also revealed that it was the search engine’s third most important ranking signal.

RankBrain is a self-teaching (aka ‘machine-learning) system which Google uses to interpret the meaning behind users’ search phrases (particularly long-tail natural language queries) so the search engine can provide relevant results which don’t necessarily contain any of the words in the query.

To do this, RankBrain processes past searches (and their wider context, such as the user’s geographic location), spots connections between seemingly unrelated keywords and phrases, and groups the data into ‘vectors’ it can understand. This way, it can understand the wider context of each search query and get a better idea of what the user actually wants to find.

When it comes across a term or phrase it hasn’t seen before, it can make an educated guess at which other terms might be related to that query, and use those terms to inform the search results the user receives.

Google has been building connections between keywords to provide better results for a while, but it’s understood that they were usually built manually by the human team behind the search engine. With the arrival of RankBrain, Google can process connections much more quickly for a much wider pool of references.

How can I use RankBrain to boost my search ranking? Right now, RankBrain is still shrouded in mystery, and no one is really sure how it functions as a ranking signal. What is clear is that RankBrain is yet another signal of Google’s commitment to improving their user experience and understanding what people want to find – not just matching results to keyword strings.

To perform well for SEO in 2016, you’ll need to research the real intent behind users’ searches, and provide a web presence that caters best for their needs.

Knowledge Graph

Back in 2012, Google rolled out Knowledge Graph – a database of knowledge collected from various authoritative sources, which allows users to simply type a question into the standard Google search bar and get an answer read back to them without the need to click on any of the search results.

The Knowledge Graph helps Google to recognise ‘entities’ – be they people, places, animals, events, media, objects, concepts, etc. – and the associations they have to other entities and facts.

This gets more interesting when you add in conversational search. Let’s look at an example; click on the microphone icon in the search bar to bring up Google’s voice search, and ask it ‘who wrote Frankenstein’. In this case, Google’s Knowledge Graph will tell you the answer is Mary Shelley – it recognises the term Frankenstein as a book, it understands that you’re looking for the author of that book, and it recognises the author as Shelley.

Now, open the voice search again and ask ‘when did she publish it’. In isolation, this query is meaningless – ‘she’ and ‘it’ could refer to anything. However, Google is smart enough to know (based on your previous search) that ‘she’ refers to Shelley, the ‘it’ refers to the book, and will understand you’re looking for the date of the book’s first publication.

You can even continue the conversation further with additional questions. Try asking ‘who was her husband’, for example!

How can I use Knowledge Graph to boost my search ranking?

If you want to appear in Knowledge Graph results, first you need to promote your website as an authority on the subject in question. Check out Schema.org, which hosts various data markup schemas to help Google (and other search engines) understand your website’s information more clearly.

Be aware that Knowledge Graph results can actually harm your web traffic – if Google can provide all the answers your users are looking for from within the search results page, they’ll have no need to actually visit your site!

PageRank

PageRank is the oldest and perhaps the best known of Google’s various search algorithm iterations, designed to rank results based on the number of links pointing to each page (and the quality of those links).

When Google first launched their fledgling search engine back in the late 90s, even the leading search engines of the era found it difficult to gauge the quality of the sites they indexed. Instead, site links were organised in directories and indexed based on topic. To get your site listed on the more important directories (such as Yahoo), it had to match with their criteria for assessing the quality of your site.

PageRank operated on the logic that links counted as votes of a page’s value to users – if lots of webmasters were linking to a particular site, it usually meant that site was an authoritative, high-quality result.

This revolutionary new system helped Google rise to the world’s leading search engine – but it wasn’t perfect. This emphasis on links encouraged webmasters to build links at all costs, opening the door to link spam and shady link networks. Google would eventually release the Penguin algorithm update in 2012, which aimed to penalise websites with high numbers of spammy links.

These days, PageRank is still around, but it’s just one of the hundreds of factors Google uses to assess and rank the authority of the sites it indexes – and the ability to check your own PageRank score using the Google Toolbar was removed in March 2016.

How can I use PageRank to boost my search ranking? As you might have guessed, PageRank is essentially dead – and yet, there are still blackhat SEO practitioners out there who will try to convince you that building tons of bad-quality links will help your site’s ranking.

The best way to take advantage of PageRank in 2016 is to simply ignore it. Trying to manipulate Google’s ranking system for your own game is not only morally questionable, it’s pointless – eventually, you’ll be penalised for it. Instead, focus on providing valuable content that will earn natural links from real web users.