If you search for anything related to voice search you’ll find no shortage of predictions. Such as “50% of all searches will be voice searches by 2020” or “(insert year here) is the year of voice search”.
Here at Upp we love new technology and ways to make b2b customer journeys better, and I don’t want to sound too cynical, but the thing with the future is that we can’t be completely sure what is going to happen. Don’t get me wrong, predictions are great, they help our thinking and can guide our focus in research and development but some predictions can be a bit wild.
Will 50% of all search be voice searches by 2020?
The idea that half of searches will be voice searches by 2020 is most often attributed to ComScore. It is a headline for many a post from the last few years and I’ve seen posts that even talk of pinning entire strategies on it (at the time of writing this I don’t recommend this). I did some searching and found a very good post by Rebecca Sentence at Econsultancy where a few details are put into perspective.
In her post, Rebecca discusses the “50%” prediction, in fact, came from Andrew Ng when he was Chief Scientist at Chinese tech company, Baidu in 2014. The prediction was actually that “at least 50% of all searches are going to be either through images or speech”. The inclusion of image search in the prediction really muddies the waters on the growth in voice search.
Rebecca discusses the breakdown of voice and image search and with some logical assumptions suggests that for voice search to reach 50% of all searches, there would need to be an additional 1.3 billion voice searches per day which is a pretty tall order.
While the exact growth rate is unclear, what is clear, is that that voice assistants are growing in popularity. Amazon reported that is has sold over 100 million devices with “Alexa” built in, and reports indicate huge growth in sales of Google Home systems. Furthermore, most smartphones ship these days with a built-in voice assistant. Given the growth in demand for voice-enabled devices, it is fair to predict that as the uptake of voice assistants grows, the use of voice search will surely grow, and B2B marketers watch these changes, in particular, changes to Search Engine Optimisation (SEO).
How Voice Search is changing b2b SEO
SEO has long been a cornerstone for b2b marketing, it’s a way to connect customers with answers to their questions at the exact time they’re searching. The changes to SEO over the last decade are well documented and voice search is changing the landscape further, providing both challenge and opportunity for marketers.
In 2011 we saw Apple launch, Siri, along with the rollout of Google Voice Search. By 2016, Google was reporting that 20% of searches in the Google app were voice searches showing real growth in uptake. Deloitte has also reported that over 80% of respondents in a survey had ready access to a smartphone in 2018 and most smartphones have voice search capabilities meaning a huge percentage of people now have access to voice search capability.
In 2013 Google also rolled out its Hummingbird algorithm update which places emphasis on natural language and aimed to bring context and user intent to the fore of search results, and this applies to traditional text-based search and the developing voice search arena. Without going into too much detail on Hummingbird, the key takeaway for B2B marketers on search results is that when using Google’s voice search your potential customers will only hear the top search result. This makes it even more important to be at the top of Google’s results.
How can you optimise for voice search?
Understand user intent
User intent gives you insight into why a person searched and what they’re looking for. It could be they want to buy something or are looking for information. Some search terms carry more intent than others. For example “Buy…” is fairly indicative that someone wants to buy something. On the other hand “what is…” or “how…” are indicative that someone is looking for information. Sometimes, however intent may be less obvious.
Google Hummingbird means that even if the intent isn’t immediately obvious, Google will still look into the context (location, device etc) of a user’s query and will provide its most relevant answer.
With this in mind, you should seek to provide answers to questions with your content. Use natural language and provide details. The number of words in your posts is less relevant than the answers you’re providing. That said to provide enough detail on a topic you’re likely to need at least a few hundred words.
Long tail keywords
When typing out searches we’ve become used to using snappy phrases that would sound robotic if spoken out loud such as “b2b marketing advice”. That isn’t how voice search works, people tend to speak as though they’re having a conversation, so “b2b marketing advice” becomes “where is the best place for me to get b2b marketing advice?”. The expectation also seems to be that voice assistants accommodate this conversational approach.
Answering questions is where long tail keywords within content are key. Long tail keywords are those which contain multiple words (3 or 4+). They are useful not only for voice search but for your SEO efforts in general as they often face less competition and top rankings can be easier to achieve. Focus your content on longer tail keywords and phrases and include synonyms and similar words to build context and help search engines understand the intent of your content.
Finding long tail keywords requires an understanding of user intent mentioned above. Armed with this understanding of the type of questions your target customers are looking to answer you can use tools such as answer the public to uncover specific questions people are asking, enabling you to create content that answers those questions.
While not a tip for voice search specifically, page speed is an important metric that extends into the whole user experience of your website. It is increasingly important so well worth improving, regardless of whether you’re optimising for voice search or for conversions.
A key reason people use voice instead of typing is that it’s faster. We speak on average 3 times faster than we type and the logical thought is that someone using voice search is likely on the go and wants an answer quickly. Your site must match expectations if it is to engage people.
To begin with, analyse your page speeds using tools such as Google’s PageSpeed Insights or GTmetrix to understand and benchmark your performance. Google and GTmetrix will tell you the speed of your pages and also give guidance on how you can improve load times.
Optimise for ‘Featured’ snippets
As mentioned above, Google will feedback its top result for voice queries, so it’s important to appear in that top organic spot. Featured snippets provide users with a direct answer to their query and ahrefs studied 112 million keywords and found that 14 million of them had a featured snippet in the search results which is a sizeable number of keywords.
To optimise for featured snippets, bring together your understanding of user intent and long tail keywords and questions and create detailed content that answers those questions.
Finding your voice in a changing b2b marketing landscape
B2B lead generation is pretty complex with lots of routes to customers. Voice search is one of the latest of these routes. Unlike other brand new channels, voice search doesn’t require any revolutionary new technology, more a different way of doing something (SEO) that has been around for many years already. I suggest bringing voice search optimisation into your plans as part of your wider marketing plan. The idea that 50% of all searches will be voice search by next year is somewhat ambitious but there’s no doubt that voice search is growing. B2B brands need to keep up with the change by adapting their approach to creating content. At the core of your content strategy, you should focus on understanding user intent and creating content which is conversational and answers specific questions.