psdesign1 - Fotolia

IT leaders need to speak up about the voice channel

Voice assistants represent the next frontier, and a range of new skills will be needed to move beyond pilot projects to a true voice channel

This article can also be found in the Premium Editorial Download: Computer Weekly: Climate change – the challenge for datacentres

Businesses are seeing the benefits of voice assistants and chatbots to provide new ways to engage with customers, according to new research from Capgemini.

More than three-quarters of businesses (76%) said they had realised quantifiable benefits from voice or chat assistant initiatives, and 58% said those benefits had met or exceeded their expectations. Benefits included a reduction of more than 20% in customer service costs, and an increase of more than 20% in the number of consumers using digital assistants.

The report, Smart talk: How organisations and consumers are embracing voice and chat assistants, was based on a survey of 12,000 consumers and forecast that by 2022, customers will increasingly prefer to use voice assistants across the consumer journey.

Capgemini found that almost three-quarters of consumers (74%) said they used conversational assistants for researching or buying products and services. But although a large majority of organisations (74%) said conversational assistants were key enablers of their companies’ business and customer engagement strategy, the importance attached to this concept is not yet reflected in actual deployment levels.

Steve Hewett, Global lead for Retail Customer Engagement at Capgemini said: “We started doing research into conversational commerce in 2017. It will be the next major channel for retailing and service business to consumer engagement.”

According to Hewett, voice assistants such as Alexa from Amazon, Google Assistant, Microsoft Cortana and Apple’s Siri offer business a key link between human-based customer support and digital commerce platforms.

Hewett said the study found that companies are mainly playing around with voice technology. “It is mostly a proof of concept,” he added. The challenge they face in production environments is how voice assistants can be integrated with  their technical infrastructure.

For example, said Hewett, the support infrastructure is not yet there to enable a customer to say something like: “Hey Waitrose, what did I order last time? Can I order this item again?”

Another issue for organisations is that there is very little knowhow of how to develop a seamless voice-based user interface. “Humans have talked for ever, but most professionals have built things for a screen,” he said. “There is a skills gap.”

Hewett believes user interface designers will need to rethink their approach to creating new experiences for the voice channel.

The technology is not there yet there to enable the voice assistant to fully engage in a conversation, he said. However, Google’s Duplex has shown that it is entirely possible to use a voice assistant to make a restaurant reservation on behalf of the user.

In a blog posted in March, Scott Huffman, vice-president of engineering, Google Assistant wrote: “All it takes is a few seconds to tell your Assistant where you’d like to go. Just ask the Assistant on your phone: ‘Book a table for four people at [restaurant name] tomorrow night.’ The Assistant will then call the restaurant to see if it can accommodate your request. Once your reservation is successfully made, you will receive a notification on your phone, an email update and a calendar invite so you don’t forget.”

In effect, Google Duplex demonstrates how a computer could be used in the future to trigger a call and interact with a human.

Read more about voice-based user interfaces

Hewett said such technology may one day find its way into the call centre, where a voice assistant is used when people request a call back. “Voice assistants will also create a much more fluid interactive voice response system in the call centre,” he said. “Within the customer service arena, it will be able to achieve the business goal of being able to handle lower tasks that are less complex.”

For instance, at the end of 2016, Santander introduced a new service in the UK to enable its customers to use voice assistants to make payments.

Stan Sthanunathan, executive vice-president at Unilever, said: “The biggest experience we’ve had is to not look at conversational interfaces as a cure for all the problems you have, but instead to use them to augment human intelligence. This makes human intelligence a lot more productive.

“Voice or chatbots can communicate with multiple people simultaneously. They therefore help in reducing the amount of stress and strain on our human agents who are responding. These interfaces eliminate between 20% and 30% of issues reaching the human agents because they are answered then and there. And even when the issue is guided to a human being, it is actually a lot more purposeful.”

Capgemini’s study found that retailers are playing with voice and automating messaging using chatbots that integrate into their e-commerce platforms.

Hewett said: “There is not a business out there that does not believe voice will be an important channel.”

In the report, Matt Kelleher, online director at Morrisons, explained the significance of voice assistants in retail. “It is important we follow this emerging trend and make Morrisons an easier and more accessible place to shop for groceries online,” he said. “It’s exciting that our customers can now shop without even needing to log in to a computer or mobile phone.”

The idea of using voice rather than a computer or smartphone enables people to multi-task. Capgemini found that in the automotive sector, manufacturers are using voice assistants to allow drivers to control the vehicle and access concierge-driven services.

In the report, Henrik Green, senior vice-president of research and development at Volvo Cars, said: “Soon, Volvo drivers will have direct access to thousands of in-car apps that make daily life easier and the connected in-car experience more enjoyable.”

Read more on Artificial intelligence, automation and robotics

Data Center
Data Management