Optimising websites for voice search – future-oriented SEO

Google’s algorithm changes almost daily but massive fluctuations in how the ranking works are rare. However, a recent trend could have a decisive impact on search algorithms in the future: the growing use of voice search on smartphones. In order to prepare for this, you need to optimise your website for voice search.

This technology is still in the development stage: in October 2011, the voice assistant, Siri, was unveiled for the iPhone 4. This was followed by OK Google, Cortana (Microsoft), and Amazon Echo. The trend: using two-way communication to have an authentic conversation with a device, or in other words – the person asks, the machine answers. This trend also influences further development of search algorithms, changes the user’s search behaviour, and has a decisive impact on SEO.

Smartphone users were initially cautious with the new technology. Even in 2013, 85% of iOS users said that they didn’t use Siri. This makes it all the more remarkable that further surveys predict an increase in use over the next few years. Out of 1,800 participants who took part in a MindMeld survey, 42% said that they had used a voice assistant for the first time over the previous six months.

Considering the growing importance of voice search, you will need to make sure that your website is optimised for it. If possible, you should start optimising as soon as possible since Google and other search engines rate this highly. But what do we know about the voice feature and how it changes search behaviour? How can websites be optimised for voice search?

Speaking instead of typing: when is voice search used?

The advantages of the voice feature are obvious: it’s straightforward, quick, comfortable, and doesn’t require any manual input: 87% of those asked by Google see this technology as the future. It is used for general device functions (calls, composing messages, selecting songs from the media library) and for online search queries.

Which providers are most likely to benefit from voice search optimisation? Voice search is particularly popular when it comes to mobile search. It is used, in particular, to obtain directions or to find local shops. This makes it especially relevant for websites with high mobile traffic and local references.

The voice feature offers freedom and comfort since no manual input is required. Typical situations where the voice search feature is used are in front of the TV, in the car, while cooking, and even while taking a bath. In addition to social networks, relevant blogs and online guides (e.g. with cooking, building, and repair instructions) could also benefit from the changes caused by the voice search.

The mobile voice survey by Google records a heterogeneous image of the application areas and user scenarios among different generations. While teenagers often use it when hanging out with friends and for helping with homework, adults often use it while cooking or for dictating texts. Teenagers, as well as adults, use it a lot in front of the TV.

Overall, younger user groups use voice search significantly more frequently than other age groups. These digital natives are a lot more familiar with the new technology and the voice feature is particularly suited to their user behaviour (Keyword Second Screen). It’s especially advisable for Webmasters who want to reach a younger target group to optimise their website for voice search.

Voice search is being used more and more often

Second screens are becoming more popular. We are referring to using a second screen while watching TV. This increase has been witnessed in the consumption behaviour of all age groups. A study by Accenture  found that 87% of consumers use more than one device at a time with the smartphone making up 57% of this amount. Since the use of voice search is particularly common in front of the television, studies on second screen behaviour are particularly valuable for voice search optimisation.

It’s predicted that the use of voice search will continue to increase. In MindMeld’s study, 88% of respondents said that they felt 'neutral' to 'very satisfied' with the voice search. This means that the voice feature rates very highly when it comes to user satisfaction – especially in spite of the fact that the technology is still in its infancy. However, since more than 30% of the respondents said that they hadn’t used a voice assistant yet, there is still great potential to attract more users.

Impact on users’ search behaviour: the most important changes

If you would like to optimise your website for voice search, it is important to understand how much search requests can change due to new technology. The biggest changes concern the language itself: search requests are moving away from the standard written language and are moving more towards a natural use of language. This means that search requests will be longer. Instead of one to three keywords, which is usually what is typed into a search engine, voice searches consist of two to four keywords with an increasing trend. This increases the relevance of longtail keywords.

The longer search queries come about as follows: keywords are often embedded in questions due to the language use being more natural. For example, if a user wants information about a celebrity, their voice search is probably not 'Steve Jobs', but is more likely to be 'Who is Steve Jobs?'. These so-called W questions come up time and time again during the voice search so they need to be taken into consideration for voice search optimisation. Question words like 'who', 'when', and 'where' are used much more frequently than 'what', 'why', or 'how'. The voice assistant also understands prepositions (e.g. 'from' or 'to') thanks to algorithm adaptations, so it is a lot more accurate when you search for 'flights from Los Manchester to Berlin' than it was a few years ago.

This further development also affects traditional text searches: the structural shift from keyword to content. The classic money keywords are losing search volume, and semantic contexts are becoming more important. This is probably because users are showing more intent with their searches. Studies of user behaviour also show that the voice search is used to acquire information – more specific inquiries and purchasing decisions are made later. In particular, content that provides specific information and answers search requests will be more successful in bringing users to your website rather than the individual keywords or a specific product.

This new search behaviour, in turn, has an impact on the technology since its algorithms are constantly being developed. It’s important to meet the user’s need for direct answers to their W questions. This is where Google’s Knowledge Graph that was introduced in 2012 comes in. The Knowledge Graph shows several search results in a separate area above the common search results; these provide graphics and a small collection of facts on the search term, without the user having to go to another website for this information. Google is developing into an independent answer machine. Users are often able to receive these 'direct answers' as a spoken response e.g. by means of the voice assistant, Google Now. These are steps towards natural two-way communication between humans and technology.

The popularity of these direct responses is problematic from the web master’s point of view since it reduces traffic normally caused by Google searches. It is, however, questionable whether the users who prefer direct answers are also visitors who would be interested in analysing them deeper on a more long-term basis.

An additional technical development complicates effective search engine optimisation: newer algorithms are increasingly aiming for personalised content in order to provide even more personalised search results and customised ads. For this personalisation, personal voice assistants provide detailed information. They are interfaces between the user, his or her device, and what they search for online, meaning that they are able to gather diverse user information. It is difficult for SEOs to analyse and calculate personalised content since information about voice assistants isn’t ascertainable for marketers and website operators.

Google has also picked up on the trend of local voice search. The search algorithm update, Pigeon, which was introduced in July 2014, aimed at improving the visibility of local businesses. The voice search is three times more likely to display local results compared to typed search, showing that Google is already reacting to mobile use regarding voice search. Make sure you consider these technological developments in your voice search optimisation.

The current voice assistants at a glance

In order to optimise voice search, you need to first deal with current voice assistants. They are designed to help you navigate and operate the device (app access, music playback, creating messages) and organising personal content (calendar, reminders, birthdays, etc.) using speech recognition. In addition, the prototype Siri was already designed to provide answers to general questions. Voice assistants are also used as an answer tool and therefore voice search engine. As an interface between the user, the device, and the web, they have great AI potential but are often viewed critically in regard to data protection laws, which is why many users still prefer the traditional way of manually searching.

Apple’s Siri meant that a voice assistant was accessible to a wider audience for the first time. It was introduced in 2011 as a component of the iPhone 4s and is now also available for iPads and the iPod Touch. The application starts up automatically with the words 'Hey, Siri' and was created to be a mobile assistant.

OK Google is Google’s voice feature and a central component of the digital assistant, Google Now. This is an app for Android, but recently it also became available to iOS devices and is the direct counterpart to Siri. The application responds to the words 'OK Google' - then you can give it a command or ask a question. Using the Chrome browser, OK Google can now be used on home-based devices as a feature of the regular Google search.

Cortana is the (relatively young) voice assistant from Microsoft. Initially introduced for Windows phones, Cortana has been firmly integrated into the operating system from Windows 10 onwards, making it accessible to a wider audience. The fact that Cortana is primarily an application for home-based devices makes Microsoft’s voice assistant different from Siri or Google Now. Cortana is now available for Android and iOS mobile.

In 2015, Amazon followed suit with its own voice assistants: Amazon Echo/Alexa. Only designed for home use, Amazon Echo is sold as an individual device in the shape of a speaker. Using voice commands, it plays music, works as an alarm clock, answers questions, and helps, of course, when shopping on Amazon.

Optimising your website for the voice feature: a checklist

Voice search optimisation is a future-oriented field of work that aims to give your website a better ranking. We have created a checklist to help you optimise your website for voice search.

✓ Semantic evaluation instead of money keywords

Semantic connections are becoming increasingly important for successful SEO. This is particularly true with website voice search optimisation since voice search requests from individual keywords develop into semantically more complex search phrases. Improved semantic readability on your website can be achieved by placing value on structured data. Internal links also help the search engine to better understand how your content fits together and therefore lead to better website indexing.

You can increase your site’s relevance by covering a wide range of topics. All aspects concerning your offer should be integrated into your website. Prioritisation is crucial: the most important topics, questions, and answers should be mentioned in a concise, yet well-informed way, and covered with search-relevant terms.

✓ Work with Schema.org and other data structuring services

Integrating Schema.org can help you make semantic connections more machine-readable. The markup language was jointly developed in 2011 by Google, Bing, and Yahoo! in order to standardise semantic annotations of websites. It provides standardised schemes for structuring data - the basis of the semantic web. With the Schema.org markup, you can store additional information in the HTML code to enable the machine to read out semantic data.

According to a study by Raven Tools, 80% of websites analysed are missing Schema.org data. This means that Integrating Schema.org offers considerable potential to stand out from the competition and makes your own website more visible in the voice search.

Tip

In our guide, you will find further information and detailed tutorials on structuring your data with RDFa, Microdata and JSON-LD.

✓ Give answers

Using question words is key to voice search optimisation. In order to comply with the frequent searches using W questions, you should integrate as many questions as possible into your content. Include all of the questions that you think might be posed relating to your offer. Make sure you include all answers on your website – including search terms. The classic FAQ structure can serve as orientation. One way to determine which questions might arise is to look in forums.

It is also useful to create custom ad groups or campaigns for voice searches that take question-based keywords into account. Note that some question words are used more frequently than others. In addition, the provision of microdata helps voice search users quickly find the right answers to their questions on your website.

Since Google is able to provide increasingly direct answers, it is recommendable to offer your own content for integration in a Knowledge Graph. This also increases the likelihood of being found by the voice search.

✓ Use local references

The user is more likely to use the voice search feature on a mobile device to obtain directions or find out local information. Local optimisation also helps if you want to optimise the voice search. Local-based companies are particularly challenged by the transition, but also benefit greatly from the growing number of voice searches.

Ratings and customer reviews are crucial for successful local SEO as well as voice search optimisation. Regularly update your contact information and opening times on Google or YELP. It’s often worth listing your business in other local directories as well.

Customers are always thankful for clear and understandable directions. If you mention nearby landmarks, places, or other regional checkpoints, this is especially effective for voice search optimisation. Experiment with other ways to place local features as keywords. Users who want information about a region or city would then be more likely to come across your website over your competition’s.

Note

Bear in mind that the voice assistant Alexa (Amazon) generates its search results for local providers from the assessment portal, YELP.

✓ Adapt keywords

Even if keywording isn’t the focus of voice search optimisation, there are some things that you should bear in mind. If you want to optimise keywords for the voice search, adapt them to spoken language and place the most important question keywords in prominent places – in the H1 (main heading) or H2 (subheading). In general, your website will be more visible when you use terms, phrases, and sentences to focus more on spoken language rather than written language.

Optimise longtail keywords so that your website can be found more easily through voice search. When searching for texts, users often leave out filler words, but it is still useful to include them for voice search optimisation. If your keywords match the natural language, you increase the probability of appearing in the voice search results.

Note

There is always the risk that brand names might not be properly pronounced or incorrectly understood by the voice search. Problems occur especially in regard to language games, acronyms, or nouns which are not English. In this case, SEO will be a challenge for the voice search. The only hope is for voice recognition to rapidly improve and for these problems to be taken into account.

✓ Keep your eye on the ball

Voice search is still at an early stage in its development and will continue to change, both in the way users see it, as well as in terms of technological adjustments and innovations. Meaningful voice search optimisation is therefore only possible if you are aware of the developments in the corresponding technologies and user behaviour. Keep in mind the earlier advances made by Google, such as Direct Answers or RankBrain, which are aimed at understanding semantic contexts and integrating artificial intelligence. But also the changes in consumer behaviour – using a second screen, for example – are important for voice search optimisation. Optimising web content for voice search is, therefore, only possible if the rapidly changing technical conditions are taken into account.