With the rise of Siri, Alexa, Amazon Echo and Google Home, voice search has been promising a revolution in our digital behaviour at home and work for some time now, but has it really happened?
Voice recognition has come a long way since the early days.
Remember Microsoft’s CEO Satya Nadella’s excruciating Cortona demonstration, where he asked the voice assistant to retrieve some data and she thought he wanted to buy some milk?
Remember everyone shouting furiously at Siri for weeks on end, as we tried fruitlessly to see if it was going to rain?
Five years ago there was only 75% accuracy for voice recognition; now, the best systems operate with 95-97% accuracy. This is the same strike rate as for some humans and, better, in some cases.
According to a much quoted Gartner figure, by 2020 50% of all internet searches will be conducted via voice. Right now 39 million Americans have an Alexa smart speaker in their homes, contributing to $1.8 billion of voice commerce sales last year alone. By 2022, this is expected to rise to $40 billion.
But Gartner’s predictions about the future of voice technology has also gone beyond domestic use and into the world of business.
Gartner’s figures suggest that by 2023, 25% percent of employee interactions with applications will be via voice - that’s up from under 3% in 2019. In fact, they claim that digital voice recognition technology will fully attain the Plateau of Productivity within the next two years.
In other words, voice will shortly become a properly usable business tool that will be an essential part of everyone’s daily working life.
Except that, anecdotally, it doesn’t really seem to be happening, either in the home or at work.
In fact, this econsultancy article actually demonstrates that voice technology was less used and less trusted by consumers in 2018 than 2017
And I don’t know about you, but I’m nowhere near close to sitting down at my desk at work every morning and barking out instructions at my computer.
But experts remain convinced that the voice revolution will come and that analysis, also, holds water.
The stats show that where they have been established in the home, Voice Assistants are becoming integrated into family life, they are used and useful. We are talking to them regularly and in a relatively normal way, not like we’re addressing a foreigner in a train station.
And what’s more they’re continually learning and listening (presumably with our consent), so they will only get better at their job.
So, even if we’re just using Voice Assistants to find directions, play music and pick out recipes, data is being gathered all the time about why, when and how we like to interact with voice tools most.
One day we will see the perfect storm of ubiquity, usability and accessibility that will make voice the default way we interact with digital devices for certain jobs in and out of the home.
So the advice right now is to start implementing voice search optimisation strategy on your web content to make sure you’re part of this evolving opportunity.
Luckily, optimising for voice fits well with a more general strategy to aid semantic search and deliver better, quicker and more intuitive content experiences for everyone. Long tail keywords and phrases that mimic spoken queries are recommended to create better matches for user intent and earn favourable content relevancy scores from Google.
These kind of strategies can help your content appear as snippets, which are a sure fire way to be a top result in most voice search queries. This is because when user conducts a voice search on Google, typically their device will only read back a single answer which is most often drawn from the snippet the engine has chosen to feature.
In addition, to facilitate better voice search brands are also recommended to build more FAQ that can provide rapid and highly targeted responses to spoken questions. There's more detail
And here's a great tool to help you search for long tail content, question based ideas around specific keywords
But, of course, more opportunity lies beyond search. Much of the technology being developed for voice is already fulfilling routine customer service tasks in fintech apps. Bank of America, Capital One and others, are all rolling out in app conversational interfaces to their customers.
As brands develop new Alexa 'skills' and figure out seamless ways of handing off from search to apps and websites, this technology could be harnessed in other sectors and for other scenarios, too.
With the rise of chatbots helping to improve inbound experiences in the B2B world, the possibility of using Voice Bots to help customers gather more information and request further content through question and answer, could help further streamline the lead qualification process.
As the tech becomes better at language recognition and begins to be sensitive to pitch, cadence, speed, so a whole sequence of search and brand interaction user journeys could be triggered and entirely fulfilled by voice.
The prospect of having a meaningful chat with your computer on a business topic is edging closer to reality.