Clara Shih, CEO of Service Cloud at Salesforce, explains Salesforce Service Cloud’s three-pillar approach to automation and AI and how the technology is transforming customer service.
In the run up to
in September, Salesforce announced
. The three new features were:
- MuleSoft RPA: Bots that can work across applications and legacy systems to “process documents, enter data, or take action on the user’s behalf, all without code.”
- Einstein Document Reader: Enables Einstein to “scan documents such as drivers licenses and I-9s, and take action on that data with just a few clicks in Salesforce Flow.”
- Digital Process Automation: New “drag and drop tools” customers can use to “define automation rules and logic, extract information from documents and integrate into automated workflows.”
, I had a chance to talk with Clara Shih, CEO of Service Cloud at Salesforce, about how the cloud-based software company sees automation and AI transforming, and actually humanizing, customer service.
The following is a transcript of our interview, edited for readability.
Bill Detwiler: All right. So let’s talk automation, AI, RPA and how that relates to the Service Cloud and how that’s kind of changing how organizations approach their interactions with their customers. Because I know that automation is a large part of many organization’s digital transformation processes. So I guess maybe start with how are automation and AI in the companies that you talk to and in Salesforce’s customers, how are they using it to really provide service to their customers?
Clara Shih: Well, this is such a pivotal time for customer service right now. If you think about the major forces that are top of mind for every C-level executive team. Number one is customer experience and customers wanting faster answers, they want to self service, they want to use the digital channels of their choice. Another trend around automation. And the third is know if the easy tasks are going to go to bots and automation, what’s the future of the workforce and how do we continuously up-skill our service agents and our field techs? And so it’s really exciting to be in this world right now because it really… Service Cloud is at the epicenter of these three major market trends.
Salesforce is focused on three kinds of automation: Full, assistance and optimization
Bill Detwiler: Yeah. And so I’d love to drill down on that part of, do you see a certain segment or processes, I guess, a segment of the process of that interaction more ripe for automation and using AI as part of that process than others? So I guess, what are some things that organizations, especially service organizations are automating? What types of processes are being automated right now?
Clara Shih: It’s such a good question. I mean, the thing with automation AI is that it can apply to so many different areas. It’s sometimes as hard to wrap your head around it. So I’ll just share at Service Cloud the way that we approach our Einstein AI and automation is we have three pillars. The first is full automation. Looking for repeat processes, looking for manual steps and our new Service Cloud RPA is a great example of that, right? Instead of an agent, opening a legacy application, waiting for the system to load up and looking up a piece of data, all of that can happen pretty much instantaneously with full automation. And also what we’re doing with Einstein Automate, which is part of our Salesforce platform built into Service Cloud but with just a few clicks, no coding required, any business analyst, anyone at a company can create a full end to end automation story. So that’s the first pillar is a full automation.
The second is when you want a human in the loop and this is what we call an assistance scenario. So if you’re sitting inside of a contact center and you’re talking to a customer, you’re not fully automating the customer interaction but there’s manual steps that previously, maybe the agent had to type in a bunch of texts to look something up. Now, we can do it for them. And this is a great example, Service Cloud Voice. As the customer is explaining her problem to the agent, instead of the agent frantically typing that in, Service Cloud Voice is listening. It’s bringing like what we would have at home with Siri or Alexa, it’s bringing that into the workplace so that the agent can focus on building connection with the customer. And frankly, solving the issue much, much faster and driving kind of the suggested next best actions.
Or if you look across at an organization. What we’re doing now with incident management and swarming between Service Cloud and Slack. Instead of if you’re an insurance company and a natural disaster strikes, instead of manually creating a phone tree, which is highly inefficient and you’re sharing a lot of the same message over and over again, we use assistance to help spin up the right set of experts from across the company. And it could be spread out across the country, they could be spread out across multiple countries. You’ve got claims adjusters in the field, you’ve got your catastrophe teams, you’ve got your executives at HQ who are very eager to understand the damages and how to make sure that we get people safely evacuated into temporary housing. So all of that can happen through an assist scenario. And then the third pillar that we talk about for AI and automation is optimizing.
How do we help every service organization, every company continuously improve using all of this data that we have? Whether it’s from conversations or agent behaviors and workflows. And I’ll give you an example there. Every call center struggles with the volume and unpredictability of calls that flood in. And as consumers, we experienced this, right? When we are put on hold for many minutes or even hours, it’s because a call center is feeling overwhelmed. Well, with our new Einstein Conversation Mining, we’re able to listen to all of the reasons why callers are calling in and identify which of those really should have been self-service opportunities. And recently, in a large organization that we work with, one of the largest contact centers in the US, through working with them, we found that a lot of customers were calling in just to update their mailing address. That’s not a good reason to call in. The customer doesn’t want to call in for that reason, the company doesn’t think that’s a good reason. And so by being able to optimize and analyze that data, we were able to identify that use case.
And then the company with, again, a few clicks, they were able to expose updating your address in their mobile app, on their website, through their bots. So that customers going forward, if they wanted to, they wouldn’t have to call in for that reason. And so those are the three things that automation, assistance and optimization.
Is automation crucial to businesses adjusting to customer demands at speed?
Bill Detwiler: Yeah, I think that’s really interesting. And it leads me to my next question, which is, I guess there are a lot of customer expectations and customer needs are constantly changing. And there was a time, I guess, when people wanted to talk to a live person for most things, or maybe that was just the way we were accustomed to doing things. And that’s shifted now for people to be able to do things in a much more self-service manner. And so I think that’s a good example of that. I’m wondering if you can kind of expand on that part of it. Are there… And talk a little bit about how automation is maybe key to companies addressing those changing customer needs. And I guess, you’ve given a great example of that, but I’m thinking about it in general. Like is automation the only way basically to react quickly enough these days to needs that are kind of changing in the future?
We don’t have years to make changes. We don’t have decades to make changes like maybe we used to. And do you think automation is key to companies be able to react quickly enough to their customers to provide that really good experience that they want?
Clara Shih: That is such an interesting question. I think the answer is yes. Just because customer expectations and frankly employee expectations are sky high and you’re right, there are still plenty of instances and plenty of customers who do want to call in and talk to someone. There’s more complex issues or in financial services, if someone’s relative, loved one has just passed away. That’s not something that they want to, most of the time self service through. They want to be talked through that process. But I think that’s the power of automation is taking the none strategic repeat tasks and freeing your workforce to focus on those higher touch, proactive, more complex issues.
How do companies successfully use automation to augment their human workers
Bill Detwiler: That actually is a perfect segue to even my next question, which is one of the criticisms that you see a robotics, RPA automation is that it either dehumanizes the experience, but it also eliminates jobs from the workforce. So from what I hear you and I’ve talked to a lot of people and they’re trying to say that… Their position is that automation is augmenting the workforce and it has the potential to do exactly what you just said. Which is free up agents who used to handle the repetitive tasks to handle more complex or specialized tasks. If that’s the case, how did companies, or how did the customers that you see who are doing it successfully, what’s their secret to doing that successfully? Because like you said, the last thing that some customers want to do is talk to a bot when they’re dealing with a very complex issue.
SEE: Salesforce lays out its Cloud 3.0 strategy (TechRepublic)
Clara Shih: It’s so true. I think, let’s talk about the pitfalls. So a pitfall might be to push too many things to self service or not offer an option to have high touch. A recent example, I was just with a customer in the insurance space where when someone’s house burns down, they don’t want to talk to a bot. They want to be comforted. They want to talk to a human being. You could build a bot, but just the data shows most of the time that’s not what the customer needs at that moment. And so being really hyper aware of that customer experience, of that customer needs state and making sure that you offer choice, but also have empathy in how you guide the channel and the right agent to do the job. I think another pitfall…
The second pitfall that I see is that companies don’t provide the continuous up-skilling for their human agents. But I think that’s a pretty common in the customer, at least in the contact center industry. Where agents are almost… It’s a high turnover business and agents kind of cycle in and out. And we really have to shift our mindset as an industry from viewing agents as kind of fungible interchangeable to really viewing them as a workforce, an intelligent workforce that deserves investment in these AI assistance tools and deserves ongoing training opportunities. I’m really proud of a couple of things that we’ve done here. Both our new workforce engagement management solution has personalized training built in. And so instead of subjecting all 10,000 people in our contact center to the same eight hours of grueling training, it can be really targeted. And might identify that you need training in one particular area and that I might need training in a completely different set of areas. And we can each be on our journey and up-skill over time. And frankly, we can earn more over time as we become more skilled in more areas.
Bill Detwiler: Is that a hard argument for you to make to companies who have seen? And I will cop to being a help desk tech, way back decades ago when I started in IT. I worked help desk, I worked in a call center. Luckily, I didn’t do external support, I did internal support. So I feel that kind of frustration of a lot of people who are CSRs, and I’m curious if you think that that’s a hard argument to make to organizations that for a very long time have seen that workforce is very transitory, as fungible, as not really not really investing in the growth in that workforce, or maybe investing in one or two people to be a team lead or a manager or something like that. But there’s not really a defined career path. I mean, have you found when you make that argument, which I think is an excellent one and should be made, do you find that people listen? That forward-thinking companies really do want to do that and see the value in that?
Clara Shih: I’ve seen the conversations shift dramatically in the last 12 months by contact centers and retail companies. It’s a tight labor market. It’s hard to hire people and it’s become imperative therefore to retain the workforce that you have. And so I think the mindset has completely shifted. And we saw this happen in a lot of office jobs too, before where we went from viewing people that way to realizing that everyone can have a growth mindset. And companies and leadership need to encourage that. We’re seeing that same thought process permeate the contact center in this labor market. And also as companies realize that it’s more important than ever to service their customers because that’s how you drive that ongoing loyal relationship and growth.
SEE: Digital transformation: A CXO’s guide (free PDF) (TechRepublic)
Bill Detwiler: Yeah. And I take it that is the key part that what every company should really strive for is to have, if you have engaged employees, they’re going to provide an engaged and a good customer experience as well, right?
Clara Shih: That’s what all of the data shows and we experienced it, right. You can tell as a consumer, when you call into a contact center and someone is just, they’re just mailing it in. They’re doing their job. It’s transactional. They know that they might not work there the next week or the next month versus someone who really cares. And someone who is almost like a brand ambassador for the company. And it just leaves you with a complete night and day different feeling in terms of how you want to continue your relationship with that company.
Bill Detwiler: And automation, I guess, I mean, is a place to kind of round it up. It sounds like building those automation processes and tools and giving companies the ability to build that into their processes is key to making that transition to the humanizing, both the interaction with the customer, but also changing the career trajectory for those agents.
Clara Shih: That’s right. It’s a real career moment for those agents. And then we have to be extremely careful, as you said. Technology is a powerful… It can be used in powerfully dehumanizing ways if we’re not careful. And so that’s why I’m also really proud of our design ethics team and just really putting that agent human experience at the heart of what we do and making sure that it doesn’t… I’ve heard some technologies out there that contact centers are using, especially with agents working from home where it sounds like Big Brother. That’s not the best way to get the most out of your workforce and to build trust. And so we believe we have a highly ethical humane way where we’re trusting our people. Where we holding them accountable, but we’re treating them with dignity and respect. And as you said, that is how you end up delivering an excellent customer experience at the end of the day because those agents feel empowered, they feel trusted and they feel loyal to the company.
- Salesforce TrailheaDX 2021: One big reason why non-developers and devs should tune in (TechRepublic)
- Salesforce launches Digital 360 for Industries, expands product offerings for retail, healthcare, financial services and more (TechRepublic)
- Transportation trends: Self-driving vehicles, hyperloop, cars communicating with crosswalks, AI, and more (free PDF) (TechRepublic)
- TechRepublic Premium editorial calendar: IT policies, checklists, toolkits, and research for download (TechRepublic Premium)
- Artificial Intelligence: More must-read coverage (TechRepublic on Flipboard)