November 19, 2019

Why Salesforce Is All-In on Voice Technology

By Devon McGinnis, Salesforce Newsroom

"How do we make the voice experience customizable enough so that it meets every culture's needs, every business' needs, and helps them meet their objectives?" These are the types of questions that keep Yakaira Núñez awake at night. 

Núñez, who leads user and product insights for Einstein at Salesforce, is dead-set on delivering the best voice technology any organization has to offer. “We're creating experiences so that business customers can do their jobs better, more simply, and work more effectively. So they don't have to have the onerous responsibility of needing to key in their data — instead, they can recite it and then things are automatically updated on their behalf."


The rise of voice tech

It’s been less than five years since Amazon launched its Echo smart speaker and artificial intelligence (AI)-powered voice assistant, Alexa, but since then the world’s become a much chattier place. Supercharged by recent advances in machine learning and natural language processing (NLP), intelligent voice-enabled assistants are now used by 47.3 million U.S. adults. Because they can grow continuously smarter by learning about each individual user’s preference, voice assistants can play a valuable role in augmenting the customer experience and building deeper relationships with brands.

According to IDC, one-third of enterprises will use conversational speech technology for customer engagement by 2022. Research from Salesforce shows that already 75% of business buyers say emerging technologies (including chatbots and voice assistants) are changing their expectations of companies.

 

“We believe AI-enhanced conversational interfaces can transform business processes, which is why we are focusing heavily on innovation in this sphere and building in voice across the entire Salesforce system,” says Michael Machado, Senior Director of Product for Einstein Voice. “Voice allows companies to command and draw insight from Salesforce in a natural way, making an already powerful system even more powerful.”

But as we discovered when we set about building voice capabilities into the Salesforce Platform, it’s not enough to simply design a personable chatbot.

“You’ve got to go deeper than that,” says Núñez. “We looked at the business challenges voice can help companies solve and we realized they are based on a very familiar human challenge — how to harness technology so users can become more productive.”

Research shows that speech recognition software has now improved to the point where it’s a lot faster to talk to the AI that’s powering your device, rather than typing on it. “We realized that if we can reduce how long customers need to spend keying data into Salesforce, they can get extra time to engage with their own customers. They reduce the tasks they have to do in order to get the bigger job done,” Núñez continues. “So today, optimizing how people interact with tech to accelerate their productivity in that kind of a way has become our starting point for any offering we devise for Einstein Voice.”

 

Listening to customers

As they explored the potential business benefits of voice technology, Salesforce’s AI experts also did something else: they listened.

“At every opportunity, we ask our customers what they like about our products — and what else they want,” says Machado. “Over and over again, we were hearing that they wanted to be more hands-free, leveraging voice-based interactions to help them do their job more simply, and work more effectively.”

Salesforce heard, for example, from service agents who wanted to use voice as a more effective way to coach call center reps. “They told us that instead of having their reps watch a training video or listen to a recording, a roleplay situation with a voice-enabled tool would be more powerful, helping their people gain hands-on experience and practice for real-life situations with customers,” Machado says. 

“We also listened to marketers who wanted to use a voice tool to gain faster, real-time insights into how their marketing channels were achieving campaign goals. And we heard from sales reps out on the road, who wanted to do things like set meetings and create follow-up tasks that updated automatically, without always needing to stop what they were doing to type.

“One thing that everyone was saying was simply this: ‘We want to Salesforce to make us more productive.’ So we realized we needed to think big, push the limits of voice technology and build offerings that would meet this need and more, across every industry.”

 

Improving the experience for sales and service

The Einstein Voice team worked to match the convenience of leading consumer applications in products that could also meet a huge range of business needs at scale. That included enabling conversational updates on the Salesforce platform — anything from ‘how’s my forecast looking?’ to ‘help me create a new account for the contact I just met’ — and putting them to work with AI. It also included enabling Salesforce’s customers to build their own voice-activated assistant, brand it for their company, and engage the end-customer on any smart speaker.

“Using these features can usher in new levels of productivity,” says Machado, who recalls a customer telling him about a top-performing rep they had who always closed a lot of deals and logged a lot of customer meetings. “They had an issue with the fact that they never had a lot of insight into what this person actually discussed with customers — they just had to rely on the numbers.”  

Conversing feels easy and natural to us —­ it’s how we communicate, and share knowledge and emotion. By freeing people up from the task of keying in data, voice can give them an easier way to enter vital details about their activities into the company system.

“Humans were born to talk, not type — voice supports a natural inclination to divulge more pertinent information as we speak,” says Machado. “In this case, the additional information gleaned helped my customer better understand how to follow up on their rep’s deals and deliver solid customer service.”

 

Delivering customization

Salesforce’s AI experts work closely with the Einstein Voice team to ensure that their offerings incorporate the latest advances in machine learning and NLP. While we are not yet at the point where voice assistants can do everything in the way that humans do, it’s these advances that are making voice-enabled devices more capable of handling nuance in conversation, and more readily adaptable to different contexts.

Such advances also help ensure that voice-enabled products can be customizable across multiple platforms and can extend across multiple boundaries — from device type and workstyles to industries.

“Delivering an accurate customer experience is incredibly important to us,” says Machado. “The way a doctor talks to a patient is obviously going to be different to how a sales rep deals with a customer — they all require a different approach to how they interact with the Salesforce system. So, delivering that customization is one of the major issues our team thinks about on a daily basis.”

 

Making an impact on broader business outcomes

The Einstein Voice team also worked to ensure that their voice tools help businesses get the right information at the right time, so they can execute against it.

“Let’s take an example from the sales side,” says Núñez. “Typically, an agent would look for a voice assistant to help optimize the way that they engage with their customers, so that they are intelligently informed about what needs to happen next.”

To achieve that, agents need insights based on predictive analytics, rather than just a keyword. “Instead of: ‘You have a meeting in five minutes,’ wouldn’t it be great to get: ‘There’s traffic on the way to your meeting, so take this route instead. Oh, and for context, bear in mind that the person you’re meeting with will be digesting these news reports that show their stocks aren’t doing so well this week,’” Núñez says. “It’s all about supporting customers to help minimize risk and failure, and instead optimize success.”

 

Voice: the new frontier

In the longer term, AI and voice recognition technology will continue to evolve, offering customers new capabilities.

“Today’s relationship between voice and consumer is just scratching the surface of what’s possible,” says Núñez.

“Soon, we may start to see voice-driven AI increasingly able to recognize our emerging needs before we do. Imagine a voice assistant able to act as a call center partner, flagging when might be the best time for an agent to interject and deliver an upsell to a customer by providing just-in-time feedback. Or imagine an assistant able to act as an expert advisor at any boardroom table, contributing to the discussion with meaningful data and insights relevant to the conversation. It’s exciting because there is so much opportunity for this technology to help businesses drive positive outcomes.”

If businesses are not already primed to consider how they approach voice with respect to brand messaging and business objectives, they should do so without delay, says Peter Schwartz, Senior Vice President of Strategic Planning at Salesforce. “Otherwise, the future consensus may be that they just don’t care about [their] customers.”

But that’s not the only reason. Núñez puts it this way: “With voice, we have the opportunity to reframe the way that we engage with technology so that we are enabled by it and in partnership with it. It can truly become a force to optimize our lives.”

For more on how Salesforce brings voice to every customer experience, see the news announcement.