laptop coding
3 March 2017 | Jonathan Lyon

Machine Learning Masterclass #3: The Future of Machine Learning in SEO and Social

Machine Learning has become an integral part of day to day life, particularly in the digital world. Companies such as Google, Facebook, Netflix, and Amazon now use Machine Learning to continually update their services and algorithms. Google is using Machine Learning to qualify their traditional ranking algorithm with RankBrain, as we have already discussed in our previous blog.

However, Machine Learning is going to impact digital beyond the reach of Google’s RankBrain. In theory, Machine Learning can ultimately change the daily life of digital marketeers. So let’s take a look at what the future may hold.

Identifying solutions to audits

The basic definition of Machine Learning is an algorithm that allows a machine to teach itself using a large set of data. This opens up huge potential for revolutionising the way in which SEOs work.

One of the key reports for an SEO is a site audit. This allows an SEO to diagnose and measure the problems and performance of a website. But Machine Learning can take this a stage further. Machine Learning could allow an SEO to not only diagnose issues, strengths and weaknesses but begin to suggest solutions to these issues.

A Machine Learning algorithm could pull together huge amounts of previous data and server logs to provide accurate predictions as to the most significant reasons for page issues. Machine Learning would essentially allow SEOs to generate accurate SEO recommendations and gain deeper insight into what is actually making the difference in rankings.

Implementing emotional insight

Recently it has been reported that Google has been working on a Machine Learning algorithm that can rate how “toxic” online comments are. They can then use this data to moderate the comments sections of websites.

This could be great news in the world of social. On the one hand, such an algorithm points to a friendlier online experience. On the other hand, the technology allows a machine to not only understand text inputs but to gain insight from them. This opens the door to deeper emotional insight within comments and reviews.

In social campaigns, deeper insight into emotion or sentiment can enable much more direction as to which content or types of content perform better and their audience enjoys most. This could also have implications for SEO. If algorithms are drawing more data from text then Google could use this to improve their SERP features and crawlers, with a particular focus on reviews and rating web pages.

 

facebook likes and comments

 

Social engagement predictions

Datatonic, a team of data experts, posted a blog recently regarding the use of Twitter streams as an input for Machine Learning algorithms. It opens up the possibility to use a machine to analyse previous tweets and use that data to predict the potential of future tweets. For example, it could look at a set of previous tweets and discover that tweets with images gain more engagement and then predict the level of engagement for your future tweets before you even publish them. This sort of insight would be impossible for a human to quantify.

This is where Machine Learning comes in. An algorithm could be used to analyse a much larger set of tweets and gain insight into more elements of tweets. Does using a particular type of content, an image or video, draw more engagement? Are there certain keywords that tend to get retweeted more often? What time of day do these elements perform better?

A machine-learned algorithm could get close to an accurate prediction on a tweet, or social media post in general, that would be a huge advantage.

Conclusions

Machine Learning algorithms have huge potential in SEO, including automation, increasing efficiency, and reducing time cost. We have seen the world’s biggest tech companies – Google, Amazon, Netflix – take Machine Learning on board, using it to improve their user experience. It won’t be long before Machine Learning becomes more mainstream and then the possibilities for SEOs are endless.

If you are curious to know how we see the future of Machine Learning, its connection to digital marketing and its potential SEO applications, then keep an eye on this blog series where we’ll be exploring its practical applications.  If you’re interested in how Machine Learning can affect your SEO rankings talk to one of our technical SEO experts today.

 

Jonathan Lyon