Artificial intelligence, such as machine learning, is becoming more prominent in the search industry.
Join our conversation as we discuss how it’s affecting us right now and what things you must consider to optimise and future-proof your site.
About Gabriella Galluccio:
- SEO & content manager at Robert Dyas
- Previously worked for PiDatametrics
It’s important to not confused machine learning with artificial intelligence (AI). Machine learning is a part of AI and is based on an algorithm, learning more without humans inputting that information. AI is the general concept of a machine doing a job, usually in a quicker way and better way that doesn’t involve a human. This includes self driving cars, working out the best way to win a chess game etc.
When it comes to SEO, RankBrain can be seen as Google’s first (confirmed) step to using artificial intelligence in search. Jon Earnshaw calls it the disambiguation engine. The connections between categories and topics is deciphered by the search engine and so, some may even say it is making connections like a human brain. It mimics our understanding of information and thus allows the best results to be surfaced to match the search.
AI is playing a massive role in voice search, which is going to be 50% of searches by 2020. Gabs recommends the below to prepare your brand and clients for this change.
- Target long tail keywords. Gabs will be using Robert Dyas’ blog to target voice search queries, in a similar way that category pages focus on the trophy terms (generic keywords with large search volume.)
- Answer Questions. Advice guides and supportive content that helps the customer during the research stage as well as the maintenance stage after purchase. This will allow you to appear in voice search if customers ask FAQs around your products.
- Conversations with our devices. When we use voice search we often have conversations with the AI where we ask around 5 questions before clicking through to the website. This is worth bearing in mind when writing around a particular topic.
SEO Data Analysis:
AI can be time saving assistants in SEO. They are able to analyse a lot of data really quickly and two examples of this are below.
- Greenlight’s Foundit! tool (formerly known as OneHydra) that looks at search volume of keywords and recommends pages to be created based around that data.
- Searchmetrics launched their new area called content experience. This is machine learning that analyses content on your site and competitors and suggests content gaps and optimisation.
Gabs suggests that AI will be used as data analysts to assist SEOers and therefore free up time for strategy and other areas of optimisation rather than be replacements.
AI in Link Building:
Reuters started using a robot to analyse emails with information which is then crawled and used to create more articles. This Wired article states that Reuters’ tool isn’t only there to help suggest content but also to write content based on the data it deems relevant, interesting and other? Similarly, Apple news scrapes information to create quick headlines.
As it’s artificial intelligence, no doubt it has a set of rules. So, as SEO’s who try to contact journalists in order to offer them content and data so that they write about the brand/client and thus link back to their sites, it’s about pitching to robots too. Remember to tailor emails and content to artificial intelligence as well as humans going forward.
Artificial Intelligence in Personalisation:
Netflix uses AI to create dashboards tailored to their customers. In the same way as Google shows results, Netflix looks at our previous history and the things we are deemed to be interested in to recommend new things to look at on their platform.
Google uses your location, details about you (if you’re logged into your email account then things such as gender and age) as well as previous browsing history to offer you results that will be most suited to you. With artificial intelligence there is no position number 1. So, be sure to tailor content to your audience and customers and know who they are!
Music credit: I Dunno (Grapes of Wrath Mix) by spinningmerkaba (c) copyright 2017 Licensed under a Creative Commons Attribution (3.0) license. http://dig.ccmixter.org/files/jlbrock44/56346 Ft: Jlang, 4nsic, grapes