There's no doubt that search assistants such as Alexa, Siri, Cortana and Google Home are on the rise, and following last month’s holiday gifting period, the amount of users who are looking to these devices is growing at a rapid pace. Although still not perfectly accurate, the current generation of assistants are much more sophisticated and accurate, meaning more and more people are relying on them for everyday search.

Unknown block type "undefined", specify a component for it in the `components.types` prop

Anyone who works in SEO knows that the landscape is constantly changing when it comes to typed-in search queries, and unfortunately for those who are already struggling to keep up with browser search, there are also many differences in how these AI driven search assistants rank your site’s URLs. (Sorry). This means that marketers and SEOs must develop new ways to tailor their wider SEO strategy for voice search – and with voice, or conversational, search so obviously on the rise, it’s even more important than ever for businesses to start making these adjustments now.

So what are the main differences between browser search and voice search? The two main differences are the focus on long-tail keywords, phrases and questions, and the clear and obvious focus on mobile.

Long-tail keywords

Google confirmed last year that 15% of daily searches are brand-new phrases/questions that have never been searched before, and this is largely down to the fact that people don’t type the way that they speak – especially when they’re searching for something online. The use of more natural speech, and more specifically, questioning, is on the rise as voice searchers opt for complete questions over shorter and more basic, search phrases used in typed-out searches. For example, rather than a user typing ‘weather London’ that same user may ask Siri ‘what’s the weather in London tomorrow?’.

But it’s not just the original search that’s a full sentence, the answer is as well. Whereas on a Google search, you’d receive a screen showing you the current weather conditions, the answer from Cortana may be more along the lines of: “It’s partly cloudy today with a chance of rain. The temperature should reach a maximum of 12 degrees Celsius.” If you’re on a mobile device, such as an iPhone, Siri will also include a screen showing you the conditions.

Mobile first

Old news, yes, but it’s surprising how many businesses still aren’t taking the importance of mobile into consideration. Over 60% of all searches are made on mobile devices, and a huge percentage of those are using voice search. This, combined with Google’s AMP (accelerated mobile pages) project, means that businesses can no longer put their heads in the sand thinking that they can continue to ignore their site’s mobile usability or delay investment in mobile.

Google’s mobile-friendly test is a quick way to know if your site will rank in mobile search results. It’s simple – if your site doesn’t pass the test, your site won’t rank on mobile. And if you’re not ranking on mobile, you’re definitely not going to rank for voice.

Unknown block type "undefined", specify a component for it in the `components.types` prop

It's also important to avoid Flash and optimise your load time by compressing imagery. Google’s Page Speed Insights tool is a great way to check and will also provide you with a list of optimisations you can make to improve your site’s optimisation for both mobile and desktop.

So what now?

Voice query research tools

Although not a voice research tool specifically, take a look at autocomplete suggestions that come up in various search engines which will give you a better idea of the most commonly asked questions around your product or service. Then, get your answers in your content – make the answer to the point, clear and easy to find by search engines. A general tip is to present short answers, and even better, use bullet points or numbered lists. Answering questions this way won’t only get your content prepped for voice search, but can also help increase your site’s visibility with featured snippets on Google such as the snippet result below for the query ‘How do I chop an onion?”.

Unknown block type "undefined", specify a component for it in the `components.types` prop

Currently, optimising for featured snippets seems to be a safe bet as Google Assistant and Google Home tend to read snippets from sites with featured snippets in what’s referred to as ‘position zero’. By focusing on the best answer for common questions, you too can work to get your site bumped up to position zero and therefore much more likely to have your site’s content and answers read out in response to voice queries.

Question your content

Examine your site’s content and ask yourself what questions your content is answering. If you want to be able to answer natural language questions, your content needs to reflect natural language patterns used to ask the questions in the first place. By incorporating more natural speech nuances into your own site’s content – easily achieved by using more conversational language – you’ll be well on your way to optimising for long-tail search. And this doesn’t just mean your blog, you should also use similar language in your product and service descriptions, trying to use keywords and phrases that are used by searchers nearer to the point of purchase.

Whilst you’re there, think about the readability of your content. Is it easily scannable, easy to understand and comprehensible? If not, fix that.

The long and short of it (no pun intended) is that voice search is here to stay and you can’t ignore its importance to your business. With this brings great opportunities, but only if your site is ready for it. Does your site answer your customer’s questions? Your FAQs page is a good place to start, so if you don’t already have an FAQs page, start there – and there’s no time like the present.

Interested in learning more, or looking for a partner to help you activate your voice search strategy? Contact our team today.


References: