Most brands have proved themselves capable of using technology for customer communication. However, when it comes to two-way customer dialogue – or conversation – technology seems to lag behind actual customer support. But voice AI may change all that.
As discussed in our previous article on AI in customer service, machine learning chatbots are being used everywhere from the frontline of customer service to on-page optimisation. Business chat apps are also thriving. But not everyone likes or can easily use these features.
Sometimes we just want a quick, low effort answer to a simple question. Other times we want the authoritative validation that comes from dealing with a real person. And as voice assistants, AI, and the hardware they are used on become more and more popular, a new, streamlined form of customer conversation is coming to fruition.
20% of search queries on Google's mobile app and on Android devices are voice-initiated.
Before we look at what voice AI is, we need to understand what it is not. And it is not those annoying voice-activated call direction services we have all dealt with at some point. You know: as soon as the dial tone connects, you now have a series of tasks to complete!
You follow the instructions delivered by the voice, tell it everything from the banal (your mother’s maiden name) to the intrusive (the last three digits of your account password) and then, maybe, if you press the correct option in time, you might get to speak to someone.
This is known as interactive voice recognition (IVR) software and it is categorically not voice AI. Voice recognition is a core part of the AI experience, but IVR is an automated system with a set number of instructions.
25% of customer service operations will use virtual customer assistants by 2020.
So now we know what voice AI is not, let us consider what it truly is. You will surely be familiar with Apple’s Siri; Google’s Duplex; Amazon’s Alexa; and Microsoft’s Cortana. Essentially, these are virtual personal assistants who ‘listen’ to voice commands and offer contextual responses. They can currently be used to do everything from tell us the weather to playing songs; as well as change the indoor temperature and respond to inane questions about trivial matters (go on, ask one how old they are!).
While they may seem, well, a bit frivolous to some (although many will attest to their life-changing properties) the fact is, they are growing in popularity and sophistication – giving brands like yours more reason to consider them as part of an overall customer success solution: both as they achieve a network effect and as the opportunities for businesses expand.
To give an example: Amazon’s Alexa started life with a thinly veiled commercial purpose – to make ordering on Amazon quicker and easier. However, developers can now create their own Alexa ‘Skills’ meaning that they can now build voice activated services on the platform itself. Microsoft’s Cortana offers similar functionality, as will Google Duplex (eventually), while Siri has a number of integrated apps.
45 million Americans have access to a smart speaker.
From a commercial perspective, voice AI’s technical accessibility means that brands of all kinds could soon create their own apps – as many are already. But perhaps the more pertinent question is ‘why would they want to?’
There are several reasons. Firstly, offering a low effort and more intuitive set of processes is a win for customer service: it makes things easier.
Secondly, voice commands save time, which is good for busy users and employees. In a similar way to chatbots, when voice AI is used in customer service, employees have more time to do other tasks – reducing costs while improving customer experience.
Thirdly, brands using voice AI can offer their customers more personalised communications. This means better overall engagement and an opportunity to become more meaningful and trusted in customers’ lives.
While the advantages are numerous, for many the level of sophistication needed to make a real impact – from both a customer and brand perspective – requires some finessing.
Should speaking to an AI be more like speaking to another human? Perhaps not. But more natural-sounding conversations would certainly aid the overall experience. Similarly, giving AI the ability to detect and express emotions and exercise empathy would clearly be beneficial – and crucial if they are going to play a central role in our lives as a virtual assistant.
Also, we need to consider the fact that, if we are speaking to a device then someone is bound to be listening somewhere. User privacy needs to be high on the voice AI agenda, as they become more widely adopted in our homes and workplaces over the next few years.
43% of mobile voice search users say it is quicker than going on a website or using an app.
As the number of digital channels expands, online commercial interactions are increasing too. Already we are shifting away from screen-based interactions to voice-based ones, and while the two will both persist for a while yet, AI will get smarter and less impersonal; and more brands will start to investigate the possibilities of voice apps.
Not sure how to get started? Let’s talk! We empower companies to deliver more meaningful experiences to their customers using conversational channels such as voice assistants and messenger services.