Opinion

Virtual personal assistants will become ubiquitous, but brands are not at all prepared for ‘voice’

Brands not wanting to be left behind on ‘voice’ need to urgently look at how they can make their data and services available to VPAs and platforms with machine-readable data on their digital assets – claims Nick Wilsdon of Vodafone

There is always a risk with any marketing trend that the hype exceeds the reality. Businesses are left wondering whether this trend will have an effect on their bottom line and how much focus they should put on it. This has definitely been the case with voice technology over 2018, with futurologists lining up to claim that this new medium will take over from the traditional web and mobile commerce.

There are four major players in the current voice landscape. Amazon has been leading the market with their Echo devices, more commonly known as Alexa. Google has released a range of home devices but more importantly, has integrations into Chrome and Android mobile OS. Apple has released its HomePod devices at the higher price range end of the market, but Siri is integrated into both desktop (OSX) and mobile (iOS)

Microsoft does not have any voice devices currently available but its assistant, Cortana, is available in the Windows operating system on desktop and Microsoft’s own mobile devices, such as the Lumia. This is a complicated, emerging market that is changing rapidly. Voicebot.ai reports that Apple’s HomePod raised more sales revenue in the first five months of device availability than Sonos did over the whole of 2017. Potential partnerships will undoubtedly disrupt market share.

In September, Amazon made the Alexa technology available to manufacturers and even installed it into a microwave as a proof of the concept. While Google has been less public about opening up its technology, you can now find its voice assistant in new Philips televisions.

Voice technology adoption will increase dramatically as we see this technology being integrated into common household devices, from TVs and fridges to clocks. It is interesting to consider these developments in voice technology against two other adjacent fields of technology, the ‘internet of things’ and robotics.

Our homes are becoming smart, from light bulbs to kettles. Voice technology is the natural way to interact with these smart devices, as well as home robots such as cleaning devices. Tony Prescott, professor of cognitive robotics at the University of Sheffield, in the United Kingdom, and director of research at consequential robotics, highlights the importance of social robots.

One in eight of us already has a social care need. Rather than thinking about robotics in terms of humanoid devices, social care robots will come in the form of intelligent moving tables or robotic pets. It is not hard to see how the technology behind robotics and voice technology will soon converge

The low cost of voice technology will rapidly push adoption, much as the sub $50 smartphone has for mobile penetration. This cost point allows manufacturers to integrate this technology easily, giving them a smart (IoT) product that appeals to consumers. Gartner predicts that 75% of United States households will have a smart speaker by 2020. I would argue this is a conservative prediction now that the technology is being distributed in raw form.

Voice represents a new interface to the devices around us but importantly, also with the Internet. However, while hands-free control will remove friction for some interactions there are certainly many activities that cannot currently be served through this medium.

A usability study from the Nielsen Norman Group indicated that voice-only or screen-based intelligent assistance work only for very limited, simple queries that have simple, short answers. More advanced tasks such as product comparison and shopping are difficult to conduct through this medium.

Users become annoyed when listening to an assistant talk at length about product options. There was no easy way for users to go back-and-forth and compare product options. They had to commit all the information about each alternative to their working memory in order to progress to the next action.

It is clear that we are in the early stages of developing these voice interactions to be useful to consumers and thereby generating significant e-commerce value. As the Nielsen report suggests, voice interactions are currently simple, they only require one step or simple action to complete. The technology needs to mature to allow multi-step tasks that require several stages to complete the process, such as placing an e-commerce order.

Some context

The evolution from simple actions to multi-step tasks will require voice technology to have a better understanding of context. This is a familiar problem to search engineers and those working in robotics. Users increasingly expect search engines to understand the context of their query. That context may be an understanding of their previous queries as they narrow down their search criteria or personal data to deliver more relevant results.

Both search and robotics are tackling the problem of physical context. Google reports that 43% of mobile queries involve geographic intent, while robots need to have an understanding of the physical space around them and even the emotions of the humans they interact with. Interestingly Amazon has secured a patent that allows them to offer relevant advertising based on the tone of the user’s voice i.e. if Alexa detects the user has a cold then s/he can offer to purchase cold remedies as follow-on interaction to their original query.

It is likely we will see rapid advancement in this understanding of context. And perhaps we are being too hard on this emerging technology. In 2000, studies by Nielsen Nelson Group showed that only 61% of users achieved their aims when visiting a website. By 2010 this figure had risen to 78%. Similarly, mobile usability has improved dramatically over the last three years.

There are significant advantages in hands-free interactions. Think of the busy parent looking after children and remembering to add a product to their online shopping list. The driver who will use hands-free interactions rather than risk a penalty for using their mobile phone. The productivity junkie who gets reminders on events and adds items to their to-do lists. My household particularly appreciates the Alexa Skill that records whether or not the dog has been fed yet, thus providing useful and timely information.

The challenge for businesses is to consider how voice can augment and enhance their current channels rather than replace them. What simple actions can they support to enhance their customer experience? These range from allowing customers to know where their deliveries are, answering common support queries to sending billing summary PDFs to their mailbox. There are many opportunities to reduce friction and improve brand loyalty through simple actions while we work on supporting the multi-step activities that are to come.

An important factor in the development of these multi-step activities and the understanding of context will be the advancement of personal assistants. Once Google assistant, Alexa, Siri or Cortana become smarter, they will allow more advanced interactions to happen through voice. Consider Alexa interacting with your calendar, understanding that your partner’s birthday is coming up and asking if you would like flowers to be delivered to your home address. Google gave a demonstration at its I/O conference this year where its assistant autonomously booked a hairdressing appointment.

In the same way that Uber made taxi rides more affordable and available to us, voice technology will make the personal assistant ubiquitous. Brands need to consider urgently how they can make their data and services available to these assistants and platforms. In preparation for 2019, I would strongly recommend that brands consider how the data on their websites can be made machine-readable.

Do you have feeds and APIs for your products and services? And lastly, how are you managing your data silos to allow these interactions to happen, i.e. can you actually report to customers the location of their delivery? For many companies, this will require a revision of their current systems and processes.

Voice technology is not over-hyped. This environment is being developed at a rapid pace on the back of several converging streams of technology, from robotics to AI and machine learning. However, our current expectations for voice and hands-free interactions should be realistic. Voice will not replace your desktop or mobile activity, but it can augment and enhance these channels.

Focus on the simple actions that you can deliver now and prepare your business to support the advanced and multi-step queries that users will come to expect. This preparation should be high on your agenda for 2019.

Wilsdon worries brands are not preparing for voice and the age of VPAs

Nick Wilsdon is a search product owner at Vodafone in the United Kingdom and has worked in digital marketing for almost two decades – a version of this article was originally published on Internet Retailing

ADVERTISEMENT

SUBSCRIBE

Sign up to our free daily update to get the latest in media and marketing