Artificial Intelligence and Search Engines. Part Two: Changing SEO

In Part One of our series on artificial intelligence (AI) and search engines, we looked at how AI is changing the way the users search and how search engines function. Welcome to Part Two, where we will be taking a closer look at the ways in which the SEO industry will have to react to AI and what it will mean for your website rankings on Google.

For as long as there have been search engines used by millions of people, there have been advantages for those websites and businesses that have been able to influence the rankings in their favour. As early search engines used relatively crude methods for determining results, it was historically relatively easy to optimise a site. However, over the years, algorithms have expanded and become more complex. AI is just the latest factor in this constantly evolving process. And to understand how AI is changing SEO, we need to first understand how SEO has evolved over time.

How SEO has evolved – a brief history

Today, SEO is big business: companies are willing to spend significant portions of their marketing budget to attempt to rank above their competitors in Google’s search results. The earliest recognisable search engines emerged in 1994 and the algorithms they used in order to rank the websites were fairly basic. Factors such as how many times a website used a specific word, whether that word was in the URL and the meta data, were crucial in determining where websites would be placed. This meant that a website owner simply had to ‘stuff’ their pages and their URLs with keywords to rank well – early SEO was simple.

The first major advancement in the intelligence of search came when they began to factor backlinks into their algorithms. When Google launched in 1998 it was revolutionary in the way it ranked pages because it looked at the internet as a whole, not just the content on the one website. It was able to see which websites were linking to others, and it recognised these links as an important factor in determining the importance and relevance of a site – akin to the way that a university essay cites sources. However, initially the algorithm worked on a relatively basic premise – the more links that a website had pointing towards it, the more valuable and powerful was deemed to be.

early Google search

This led to a situation in which if websites wanted to perform well in Google’s search results they could continue to utilise keywords, but also boost their site further by building a huge number of links, regardless where these links came from. For a period of time this was standard practice. However, when Google realised that many businesses were using these sorts of underhand tactics to receive an artificially-inflated ranking, they deciding to do something about it. This saw the launch of two large scale updates to Google’s algorithm: Panda, promoting the value of high quality content, and Penguin, punishing sites with large numbers of links coming from poor quality sites.

This is where SEO became a far more complicated and delicate process, and website owners and SEO specialists had to think very carefully about everything they did to a website to ensure it wouldn’t fall foul of the new rules.

A more advanced algorithm

The fallout from Penguin and Panda was enormous, and it indicated that Google was going to be continually refining its algorithm to attempt to make it impossible to manipulate or artificially enhance a website’s position. The next major update, which was known as Hummingbird, focussed on a shift towards natural language.

While webmasters and site owners had become used to using text and content to serve a purpose (to drive sites up the rankings), Hummingbird placed a greater preference for sites that used ‘natural’ language. This meant that websites that were filled with useful and interesting content ranked higher than those that simply contained a good density of relevant keywords.

There is no doubt, then, that Google’s algorithm was evolving and becoming more advanced with each change. But at this point they all had in common that there were ideas that were programmed into the algorithm by humans. However, this changed with the deployment of RankBrain.

The rise of RankBrain

Google began using RankBrain as a factor it is search results in 2015. It is an AI system that is considered to be the third most important ranking factor, behind content and links. RankBrain uses AI to analyse words and phrases that it has never seen before – it can then make a guess at the meaning of the phrase based on similar phrases. This means that it is extremely effective at showing relevant results even if it does not necessarily understand the query.

As search has become more conversational and in the form of long-tail, complicated questions, this AI is designed to help the algorithm translate the questions into something it can understand and provide search results for. Data from previous search queries is fed into RankBrain and it is uses this data to learn how connections are made between topics. It is also able to spot patterns between searches that might appear unconnected.

This is one of the first examples of AI being used to improve search results, but this begs a question: how should website owners optimise their sites for an algorithm that is learning by itself?

This is good news for SEO!

It might seem as if the addition of AI to Google’s algorithm spells trouble for those in SEO – after all, as AI learns more about websites and what kind of content a user is searching for when they use a particular search query, it becomes much harder to manipulate or influence the system in any way. However, on closer inspection, this is actually excellent news for SEO – or, more specifically, those businesses using ‘white hat’ SEO techniques.

Reputable SEO agencies and experienced professionals already know the steps that they need to take to ensure not only that their site will rank well in search results, but also won’t fall foul of penalties under the algorithm: focus on creating the best possible content and achieving links that the website deserves.

However, it has always been frustrating for white hat SEOs when they can see competitors utilising black hat techniques and getting results without being punished. Not only will AI reward reputable and genuine SEO, it will make it easier for search engines to spot poor practice. AI is definitely bad news for those agencies and companies using underhand methods to artificially inflate their rankings.

SEO

What this means for content creation and link building

Let’s take a look at what Google itself describes as the two most important ranking factors in its algorithm: content and links. These will be affected by the rise of AI.

For example, Google’s is AI becoming better at recognising the difference between genuine high quality content and simply average, non-duplicated text. This means that those websites that create the best possible content that is genuinely useful and interesting to their audience will see the rewards. This effectively means that the best advice is to carry on with the same plan that Google has been recommending for a long time: create amazing content that answers the questions of your audience and provides value to the reader.

In terms of links, things have moved on dramatically from the early days when a link from any site would do. And yet, links remain a vital aspect of determining the quality of a website. This means that websites that focus on gaining strong, earned links from powerful and relevant sites will continue to see a benefit.

Additionally, as we have seen with RankBrain, Google is getting better at understanding search intent – what the user is trying to achieve with their search term. This comes from the AI being able to more clearly understand what a user means when they type in a query or use voice search. From an SEO perspective, you can take advantage of this by tracking how visitors use your site and drawing conclusions from the behaviour of those who convert.

How to prepare your site for AI

So, what should you do in order to prepare your website for the increasing use of AI in Google’s algorithms? The truth is that AI itself will not make any changes to the way that the algorithm operates – nor will it change Google’s priorities. The use of AI is to make it easier for Google to meet its main goal: providing the best possible search results for its users.

This means that to prepare for AI you simply need to follow the same advice that Google has been suggesting for a long time. Firstly, create the best possible content that is going to be genuinely useful and informative for the user; never has the phrase ‘content is king’ been more relevant.

Remember additionally that AI is constantly getting better at understanding natural, conversational language. This means that when you create your content you must always do it with a human reader in mind. Gone are the days that you could trick the search engine with content that was ‘optimised’ – if Google spots content that looks forced or unnatural, it will be able to tell the difference.

You also need to ensure that you are earning your links. With the help of AI, Google is getting better at noticing patterns and trends. So if you are still engaging in the practice of buying links this is something that Google will notice, more so than ever before.

Finally, it is more importantly than ever to stay up-to-date with what search engines are looking for from sites. As Google and others increasingly utilise AI it will make them more capable than ever to enforce their algorithms. So it is vital to stay ahead of the game. Working with experienced SEO professionals is crucial, as the development of AI in search is fast and it can be easy to get left behind without expert advice.

Google Analytics

We hope you have enjoyed our series on artificial intelligence and search engines. This is still very much an emerging field and an exciting part of the future of SEO. Please check back to the Artemis blog regularly as we will be updating our content which further specific developments in AI and search engines, as well as providing insight into all areas of SEO best practice.

And if your business could benefit from our SEO expertise please don’t hesitate to get in contact with us today.