- info
- Clips
- Transcript
(orchestral music) Hello and welcome to this special Cube Conversation here in Pal Alto, California. I'm John Furrier, co-host of the CUBE. We are here with Jonathan Rosenberg, CTO, Chief Technology Officer and Head of AI for Five9. Jonathan, great to see you, thanks for coming in, Thanks, my pleasure to be here. So, you've had a stellar career, certainly technical career, going way back to Lucent Technologies, Yeah. Now here at Five9, Cisco along the way. You've been a real technical guru, you've seen the movie before. This is happening, Every wave of innovation, multiple waves you've been on, now you're on the next wave, which is cloud AI, as CTO Five9, you know, rapidly growing company. Yes it is. What attracted you to Five9? Yeah, great question, there's actually a lot of things that brought me to Five9. I think probably the most important thing is that I've got this belief, and I'm very motivated for myself, at least, to do technology and innovate and create new things. I have this belief that we're on the cusp of the next generation of technology in the collaboration industry, and that next generation's going to be powered by artificial intelligence. And one of the ways I sort of talk about this is that, if you look at the entire history of collaboration up 'til now, meetings, telephony, messaging, it was about figuring out a way to get the bits of data from one person to another person fast enough to have a conversation, that's it. Once we got the audio connected, we just moved the audio packets and the video packets and the messaging from one place to another, and we didn't actually analyze any of that, because we couldn't. We didn't have the technology to do that. But now, with the arrival of artificial intelligence, in particular, speech recognition, natural language processing, we can apply those technologies to that content and take all this dark data that's been basically thrown away the instant it was received, we just have to process it and do things. And that is going to completely transform every field of collaboration. From meetings to messaging to telephony. And I believe that so strongly, I said great, that's going to be my next job, I want to work on that. And it's going to start in the contact center because a contact center is the ideal place to do that. It's the tip of the spear for AI in collaboration. And it's in a real great area, to disrupt and create innovation opportunities. Absolutely So take us through the impact, because one of the things I've observed in this industry is you have, I don't want to say mainframe client server, to go back and date myself, but there was that wave of client server computing that... And mainframe's cool again, we just call it collab now. (laughs) Exactly. So you have these structural industry waves. Take us through the waves of how we got here and what's different now and why can't the old guard, or the older incumbents survive? And if you're not out in front of that next wave you're driftwood. So, what do these waves mean, why is this important, what has to change to be successful in the new wave? So there's been this whole, like you said, these waves. So the first wave of telecommunications was like hardware, circuit switching, big iron switches sitting in telco data centers. And then that era transitioned to software, and that was with the arrival of voice over IP, and technologies like SIP, and that made it more, less expensive and anyone could do it and it transformed the industry. The next wave, the third wave, we're still like halfway through, and in some areas actually just beginning. Contact center is early here. The third wave is cloud, right, is now we're moving that software to a totally new delivery vehicle that allows us to deliver innovation and speed. And that wave has now enabled us to start the next wave, which is only in its infancy, which is AI, right, and the application of machine learning techniques to automate all kinds of aspects of how people communicate and collaborate. I think cloud is a great example. We've seen AI, which has been a concept around when I was in computer science back in the 80's, there was AI theory and the science of it has, not so much changed but computing's available, the data to be analyzed for the first time is available. You mentioned analyzing the bits is now a key part. Yeah. What does it actually mean to someone who's, either has a contact center or has a large enterprise, says, you know what, I got to modernize. How does AI fit them, what is actually going on? Right, great question. So, AI actually can solve lots of different problems. The end of the day, again, AI is like, it's the biggest buzzword, right? (Furrier laughs) It's in my title, so I'm a little guilty, right? Definitely get a pay raise for it. (laughter) But what it comes down to is really the core idea of machine learning. Which is really like fancy new algorithmic technique for taking a bunch of data and sort of making a decision based on it. And it turns out, as we've learned, if you have enough data, and you can have enough computing, and we optimize the algorithms, you can do some amazing things, right? And it's been applied to areas like speech recognition, and image recognition, and all these kind of things, self-driving cars, that are all about decision processes. Do I go left, do I go right? Is this Bob, is this Alice? Did the user say "and" or did they say "or"? Those are all decision processes that these tools can automate. What does it mean in the contact center? It means everything in the contact center. If you look at the contact center, it's all about decision processes. Where should this call get routed? What's the right agent to handle the call right now? When the agent gets the call, what kind of things should they be saying? What do I do with the call after the call is done? How should the agent use their time? All those things are decision processes and they are key to the contact center, so AI and ML are going to transform every aspect of it. And most importantly, analyzing what the person is saying, connecting with the customer, allowing the agent to be more effective. Yeah, I think this is one of the most cutting edge areas of the business and the technology, and Rowan, the CEO was talking about, you know, emotional, cognitive, cognition around connecting with customers, and data's certainly going to be a big part of that. But as machine learning continues to get its sea legs, you're seeing kind of two schools of thought, I call it the Berkeley school, hardcore mathematics, throw math at it. Then you got this other side of a, uh, machine learning, which is much more learning. It's less math, more about adaptive, and self learning. One's deterministic, one's non-deterministic. You're starting to see these use cases where, yeah there's a deterministic outcome. Right. You can throw machine learning at it, great, Exactly. Boom. Humans can curate, create knowledge, create value. Then you got a new emerging use case of non-deterministic, like machine learning environments, I could be driving my Tesla down the road, or my company's running the contact center, I got to understand what's going to happen before it happens. Right. Talk about this, what's your thoughts on this, this is in a really new pioneering area. What's your view on this? Yeah, so I think it actually illustrates a key point I want to narrow in off of what you said, which is that, a lot of these problems still it's about the combination of man and machine, right? It's that there's things that are going to be hard for the machine to predict. So the human in their usage of the product teaches the machine. And the machine as it observes, helps the human achieve mastery. And that human part, by the way, is even more important in the contact center than anywhere else. At the end of the day, you're a customer and you call up, you're reaching for a human connection, you're calling because you want to talk, you've got a problem, you need someone to not just give you the answers, but to empathize with you, to understand you. And if you go back, John, to think about the best experience you've ever had when you called up for support, or get a question answered, it was someone who understood you, was friendly, polite, empathetic, funny, and they knew exactly what they were doing, and they solved it for you. So the way I think about that is that is that actually the future of the contact center is a combination of human and machine. And the human delivers the heart, and the machine delivers the mastery. And I just noticed you're, I'm looking at Twitter, right and you just tweeted this 14 minutes ago, Yeah. The future of contact center is-- It's nice to put it in a quote. The combination of human and machine, the human delivers heart, the machine delivers mastery. I think this is so important, because unpacking that words like trust come out. Yes. Truth. Um, relationships, so-- When you ask what my experience is, it's when I've gotten what I needed, knowledge or the outcome I wanted, plus I felt good about it. Right. I trusted it, I trusted the truth. Yeah. And you're seeing that in media today, with fake news, you're seeing it with, digital has kind of almost created this anonymous, non-trustworthy, it's data. There's been no real human packaging, so I think you're, I'm hearing you, you're on the side of humans and machines, not just machines being the silver bullet. Absolutely, absolutely, and again it goes back to sort of the history of the contact centers, there's been this desire to like, just make it cheaper, right, but as the world is changing, and as customer experience is more important than ever before, and as now technology is enabling us to allow agents and human beings to be more effective through this symbiotic relationship that we're going to form with each other, like we can actually deliver amazing customer experiences, and that's what really matters. And that idea of trust, I want to come back to that word. That's like super central to this entire thing. You know, you have to, as a user you have to trust the brand, you have to trust the information you're getting from the agent, you have to trust the product that you're calling them to talk about, and that's central to everything that we need to do. In fact, it's a fundamental aspect of our entire business. In fact, if you again, think about it for a moment here. We're going to customers who are looking to buy a contact center and we're saying trust us. We're going to put it in the cloud, we're going to run it, we're going to operate it for you, and we're going to deliver a great, highly reliable experience. That takes trust too. So one of the things that, back to your earlier question, why did it come to Five9? One of the things it has done is build this amazing trust with its customers. Through its huge amazing reliability, uptime, a great human process of how we go and work with our customers. It's about building trust at every single level. So I want to put you on the spot here because I know you've seen many ways of innovation. You've seen a lot of different times, but now it's more accelerated. You've got cloud computing, you've got AI, much more accelerated innovation cycle. So as users expect to interact with certain kind of environment, Rowan talked about this in his interview, the CEO Rowan Trollope, so users want to be served on the channels that they want to be served in. So having a system that they have to go to to get support, they want it where they are, right? Yes, exactly right. So how is the future of the customer interaction, whether it's support or engagement, is going to take place in context to non-linear discovery progression. Meaning, are they going to service themselves in the organic digital space, I honestly don't want to go to a site, per se, how do you see the future evolving around this notion of organic discovery, talking to their friends, finding things out. Does that impact how Five9 sees the future? Yeah, absolutely, and I think it gets back to sort of an old idea of omnichannel. I mean this is something that the contact center people have been talking about for like, forever, like the last ten years, right? And its original meaning was just this idea of oh, you know, you can talk to us via chat, or you can send us an email, or you can send us a text, or you can call us, right, and we'll work with you on any of those. But like you said, actually what's more interesting is as customers and users move between those things, and it actually switches from reactive to proactive, right? Where we actually treat those channels as well, depending on what the situation is, we're going to gather information from all these different data sources, and then we're going to find the right way to reach out to you and allow you to reach out to us in the most efficient way possible. So you see a real change in user expectation experience with real contact. Yeah, yeah, I mean one thing that technology's delivered is a change in user expectations on how things work. And if you look at the way we as human beings communicate with each other, it's dramatically different today than it was really just a few years ago, right? So Jonathan, let's look under the hood now, in terms of the customer environment, because certainly I've seen legacy after legacy systems being deployed, it's almost like cyber security kind of matches the same kind of trend in your world which is throw money at something and then build it out. So there's a lot of sprawl of solutions out there around trying to solve these problems. Yeah. How does the customer deal with that? When they're going forward, they're on this new wave, they want to be modernized, but they've got legacy, they got legacy process, legacy culture. What's the key technical architecture? How do you see them deploying this? What's the steps that they should take in your opinion? It will surprise you not one drop when I say it's go to the cloud. (laughs) Right? And there are real reasons for it, and by the way this is going to be, I'm going to be talking about this at Enterprise Connects though, so tune in, Enterprise Connect, I'm going to be talking about this. There's a ton of reasons, five huge ones actually, about why people need to get to the cloud. And one of them is actually one of the ones we've been talking about here, which is a lot of this modernization is rooted in artificial intelligence. It turns out you just cannot do artificial intelligence on premise, you can not. So the traditional gear, which used to be installed and operated by legacy vendors like Avaya, you know, they go in, and Genesis, they go and they install thing, and it works for just one customer at a time. The only way artificial intelligence works is when it gathers data across multiple customers. So multitenancy and artificial intelligence go hand in hand. And so if you want to take any benefit from this stuff that we've been talking about in this conversation, the first step is you've got to take your contacts into the cloud just to begin building and adding your data to the set, and then leverage the AI technologies as they come out. So data is the central equation in all this, because good data feeds good machine learning, good machine learning feeds great AI. Exactly right. So data is the heart of this. Yes. So data making data in the cloud addressable seems to be a key thought. Your reaction and what are you guys doing with respect to that? Absolutely, absolutely, and this is by the way another reason why I joined Five9, that I've been speckling in here. I said alright, if the future is about AI, as I said that's what I want to do in collaboration, you need data to do that. You actually have to work for a company that has a lot of data. So market leadership matters. And if you go look at the contact center, and you go look at all the industry and analyst reports, like it made it pretty obvious like who to go to. There was like the leader in call contact center with tons of agents and tons of data is Five9. And so that's why I'm here. So building the data, aggregating data, that's one of the first things I'm working on here is how do we increase and utilize data that we've been gathering for years. And a lot of the, we've had this conversation with many customers before about silos. Yes. Silos kill innovation when it comes to data addressability. Your thoughts on that and what customers can do to start thinking about breaking down those silos. Yeah, exactly. So in fact silos have been a big part of the history, of like especially on-premise systems. One si- in fact off again, one silo for inbound contacts, and a different one for outbound, different departments, by the way, also had their own different contact centers, and then you had other tools that had other data. You'd have like a separate tool over there for CRM, and a different tool over there for WFO and WFM, and something else for QM, and all these things were like barely integrated together. In the cloud, that becomes much more natural to bring these technologies together. And the data can begin to flow from these systems in and out of each other. And that means that we have a much greater access to data and correlated data across these different things that allows us to automate all over the place. So it's this positive reinforcement cycle that you only get when you've gone to the cloud. The question I want to ask you is more customer, so I'm pretending I'm a customer for a second, and I want to ask you, Jonathan, what's the core innovation for me to think about and bring to my organization if I want to go down the modernization? How do you answer that question, what is the core innovation strategy that I should have. Actually moving to the cloud is one, but beyond that, is it just cloud, then what else? What do I need to be preaching internally and organizing my culture around? Yeah, great question. So I mean I think the cloud is sort of the enabler of many of these pieces of innovation, right? So velocity and speed is one of them, and setting up and adjusting these things used to be super, super hard. You wanted to add agent seats, oh my gosh, you're going to have to go buy new hardware, and rack and stack boxes and whatever, so even simple things like reactiveness, right? That's something that's important to talk about is that many of our customers and businesses are highly seasonal, right? We've seen like somebody showed me a graph, this was like oh my gosh, like it was a company that was doing a telethon, and they said here's how many agents they have over this year. It was like two agents, and then it shot up to like 500 agents Yeah, the phone bank. for like two days, and then dropped back down. And I'm like, if you think about a business like that, you could never even do that. So cloud is nice, but the way you talk about it, and as a IT buyer of these technologies, you talk to your business owners about reactiveness, speed, velocity, right? That's what matters to a business, and then customer experience. Yeah, one of the things, just to kind of end the segment I want to get your thoughts on, I'm going to bring kind of a industry trend that's, I think might be a way to kind of talk about some of these core problems around data. Most mainstream people look at Facebook and saying, wow, what a debacle. They used my data, they're using it against me, I'm not in control of my data. You're seeing that weaponization, you know, people are saying Alexas were rigged, so weaponizing data for bad, means there's the content, and there's context, and infrustructure, cloud. But there's also the other side which is you actually make it for good. Yes. So you start thinking about this, people starting to realize wow, I should be thinking about my data and the infrastructure that I have to create a better outcome. That's right. Your thoughts on that as people start to think about AI in terms of the business context. How do they get to that moment where they can saying, I don't want anyone weaponizing data against me, I want to use it for good. Yeah. How do they actually think about- Well, it comes back to trust, by the way, right, is that to some degree it's an uphill battle due to some of these debacles that you just talked about. But contact center is a different beast of the whole thing, and interestingly it's an area where there's already been an assumption by users that when they interact with a contact center, that data is sort of used to improve the experience. I mean, every contact center, the first thing they say, by the way this call may be recorded for training and monitoring purposes. They opt in. Right, it's already opt in, there's an assumption that that's exactly how that is being used. And this is another reason, by the way, I went to contact center, is it was the tip of the spear because it was a place where there was already permission, where the data is exactly the kind of stuff that had already been subject to analysis and a customer expectation that that's actually what was happening. The expectation was there, the ability to action that data what was missing? So now we're filling in the ability to action on all that data with artificial intelligence. And final question, what's your vision going forward as CTO and AI. What's the vision of Five9? What do you see the 20 mile stare for Five9 within context of the industry we just talked about? Yeah, yeah. So it's about revolution. I'll be honest, right? And I tell people like, I'm not like an incremental, steady Eddy CTO. Like I do things because I want to make big changes, and I believe that the contact center is on the cusp of a massive change. My boss Rowan said this, and this has been actually central to how I'm thinking about this. The contact center in the next five years will be totally different than the 25 years before that. As a technologist I say wow, five years? Like that's not very long in terms of software development. That tells me we're going to pretty much rewrite our entire stack over the next five years, and show what should that start to look like? So for me, it's about how do we completely reimagine every single aspect of the contact center, to revolutionize the experience by merging together human and machine in totally new ways. The innovation strategy is cloud and AI. Cloud and AI. And data, great. Jonathan, great to have you on. My pleasure. Great conversation. Quick plug for you guys, you're going to be at Enterprise Connect, the Cube will be there covering the event as well. What are you going to talk about there? What are some of the interactions? What will be the hallway conversations? What's your objective, what's your focus there? Exactly, so I'm going to be having my own session where I'm going to be talking about the five reasons that you may not think about to go to contact center in the cloud. I've hinted already at just one of them. I think you gave two. Well (laughing) depends on how you count it. You know, AI is clearly central, and I'm going to sort of talk about the other four. Great conversation. A lot of change, massive change happening, great innovation strategy, great mission here at Five9, great mission around changing and reimagining more change in the next five years than the past 25 years. Again, cloud computing and AI's doing it. There will be winners, there'll be losers, we'll be following it from the Cube. Jonathan Rosenberg CTO and head of AI at Five9. I'm John Furrier with the Cube, thanks for watching. (orchestral music)