In the digital age, ‘keyword’ is a buzzword that has been floating around for a while. Digital marketers spend a long time researching, adjusting, and analysing their keywords either for SEO or paid search engine marketing purposes. This means, of course, that people spend a lot of time thinking about the words they use in their content in order to rank higher in search engine results and ultimately improve their business. Sounds straight forward doesn’t it? Actually, keyword research can be very complex and many realms of linguistic study have a huge part to play. If you’re struggling to reach the right people, or your content is performing way, you might want to have a little look at some linguistic theory in relation to your keyword research and try improve your understanding of the language you are using.

Recently there have been a lot of discussions about whether voice search is changing the way that marketers need to think about their keywords or not. From a linguistic point of view, keyword research and choice of keywords are becoming ever more complex, and voice search is just one of those reasons. Spoken language has always, and will always, vary from written language. This is even more true for languages such as French, which have completely different registers for the spoken and written form. Digital communication has started to blur the line between written and spoken language. Think about chat rooms for a start – instant messaging mimics a phone call but takes a written form, so it lies somewhere in between. Some linguistics believe that online language can not be distinguished and categories in spoken or written but rather is its own form. So, does voice search really have that much of an impact?

Regardless of whether it does or not, there are other reasons behind why search engine marketing is becoming more and more complex. Google’s AI is always developing and with this development comes more of an understanding of context and other elements that help Google to understand what the user is looking for. Something to remember here is that, voice recognition does not equate to voice understanding. With all search queries, Google does not UNDERSTAND what is being asked as such but rather uses keywords (and other elements) to find what could be relevant to the terms queried. With every update, Google tries to get closer to an understanding like that of a human brain. Linguists and translators will tell you that Google translate will never, and has never, replace a human translator due to the absolute complexity of languages.

What does this all mean for search? Obviously, we’re not focusing on translation here, but a similar implication implies. Language does not stand alone. It is inextricably linked to its context, interactions, and emotion. A human brain would struggle to understand what somebody was saying without the context. For example, if you were given a passage from a novel, without knowing the premise of the novel, the characters and their relationships or any other context, you would only be able to make sense of a small amount of the information you were given. It’s a little like this for Google. They have the search query (and your search history normally), but that’s it. They have to make sense of what you want from a short sentence, or a few words. Without breaching basic human privacy rights, will Google ever be able to identify the context, emotional and factual, behind a search query in order to improve its search results? Maybe not but gone are the days when a marketer can get away with stuffing keywords into their content in order to rank higher in Google. Now you need high quality, relevant, content to rank anywhere near useful, because Google has worked out how to distinguish between a keyword stuffer and authentic content. So, what else will Google be able to do in the future? The future is uncertain but as digital marketers, we need to make sure we are identifying the potential needs of the consumer.

HOW DO WE DO THIS?

A focus on long tail keywords can always help, this is beneficial to combatting potential issues in the rise of voice search and Google’s attempt to identify context and emotion in language. Long tail keywords can also be problematic for a marketer of course due to the need to make detailed predictions about what people are searching for. This issue can be resolved using broad matches (although this could create a long of unwanted clicks!), or by forcing yourself to experiment and analyse. When you begin a campaign on Google Ads, don’t presume your keywords will work. Make sure you’re checking which search queries triggered your ad and adapt your keywords accordingly. There’s no formula to choosing the best keywords. Due to the complexity of Google, so many other factors may contribute to your ad appearing from the ‘wrong’ search term, and you never quite know what vocabulary or combination of words somebody might use for their search query. Even a team of linguists with a huge corpus would struggle to predict the vocabulary and syntax of search queries.

Always, always, focus on your customers, plural. Unless you have an extremely niche product, you will have a variety of customers and identifying them can really benefit you. What one customer might search could vary greatly to another. Imagine a 60-year-old Dad is using Google to find a the best sandwich places around, and simultaneously his 21 year-old-daughter is searching for the exact same thing. Not only will their use of colloquialisms (an informal word or slang) vary on a generational level, but both people will also have their individual idolects (individual speech habits – watch Manhunt on Netflix if you’re interested idolects!). Also imagine that the Father is from Nottingham, but the daughter lives in London. Now you’ve got regional dialects (a form of language spoken in different areas) to think about, and these things don’t just apply to voice search but also change the way people write and communicate online. So, you have to think about the potential variations in what the father and the daughter might search for. This may vary very little depending but it may vary a lot depending on the product and the synonyms available for the topic. For example, the Father might search for ‘the best cobs in Nottingham’ and the daughter might search for ‘the best subs in London’. Getting both of these variations in your keywords will help Google find you for both of these customers. Sure, Google is complex enough to identify synonyms! Don’t get me wrong. But the whole point of keywords is that you’re making it as easy as possible for Google to put you in front of the right person at the right time, so it might be best not to rely on Google’s synonym database.

As for Google’s attempt to start understanding people more, the marketer can only do what they should already be doing, and write content for the consumer rather than for the search engine. This will only get more important as Google and similar platforms develop their ability to ‘understand’ content and ‘understand’ search queries in order to match the two together.

Linguistic study can be applied to SEO in many, many ways, linguistic variations, pragmatics, and computational linguistics, and more. We will be looking at how you can apply linguistics to your content in our future blogs, so make sure to keep up to date.