The ActivateCX Podcast

Unveiling AI breakthroughs

• Frank Rogers • Season 2 • Episode 25

Share episode

Find, Buy, and Optimize your Contact Center https://activateCX.arroyo360.com/activateCXtoday

Discover the latest AI breakthroughs with Arroyo360, RingCentral and Five9. From natural language chatbots to intelligent virtual assistants, learn how AI is revolutionizing the contact center industry. Check out this compilation of guest appearances from Erik Smith, Bob Lenarcik, and John C DeBaca. 

#ringcentral #five9 #AI #chatbot #IVA #disruption_4 

I think more people are aware of AI than ever before, but really it's just an enhancement to previous tools that were already in place. And the solutions that we offer are starting to think about, okay, how do we start to include some of this new functionality and technology into it to continue improving the experience? But also keeping some guardrails in place because this is a public facing thing and there's a associated risk with that. The second kind of part one B is process automation. So we're not just answering FAQs, right? This is really now complex capability for the AI to do things like schedule and reschedule appointments, update upsell orders potentially gather all kinds of information to increase conversion. So. Really a multi phased conversation that the AI is capable of having. The third big one, and this is probably the biggest growth outside of chat GPT this year has been agent assist AI. So many of you have probably heard of this or maybe seen some stuff at a trade show but by far our biggest growing bucket of AI, how do we empower the agents to have the necessary tools to serve the customer as the conversation is unfolding? Whether that be dynamic scripting pulling information from a knowledge base, it's contextual. But again, just equipping them to have a robust conversation. And then it moves into things like AI routing and AI analytics. And how do we have, you know, have that some of that great machine learning to come in and tell us what's working, what's not working. And that's both for our agent teams, as well as the AI itself. So really the machines teaching the machines, which is. I think what we've all been waiting for. So this is when you talk about your pillars, this is fundamentally a framework that you've created with inside of RingCentral to categorize various different user stories and create maybe a path to the product. That's going to solve that the best. Yeah, I think because AI has become such a ubiquitous term a big part of our job here at RingCentral is just helping to educate our customers and prospective customers and I think that's. That's so important in everything that we do. You know, don't, don't be afraid, like go get it. It's, it's there to help you if you make it. So you told that, that story about your friend. I'll tell a story. One time I was, I was out on a business visit to a client. And as I got ready in the hotel, I was putting on my shoes. I was sitting on the bed and I'm watching the television and a morning A business show and they're speaking to somebody who is actually was driving a lot of innovation in the healthcare environment and they were providing a lot of telehealth and they were interviewing this leader of this company and he was just a very direct individual. And they said, well, what about. What about your competition? And he said, my competition is dead and they don't even know it. It was like being punched in the face with a reality of part of our ability to be sustainable. Our ability to survive is our ability to adapt. And so sometimes you have to be proactive. You don't need to be bleeding edge. You certainly can be leading edge. And I think that right now there is maybe a little bit of an opportunity to win adopter around these tech. 100 percent disruption always creates opportunity. So this dynamic of how we're going to maybe downsize the organization redirect people into other parts of the organization that may have need. All of that is fights against a command and control structure, which is pretty traditional. How do you, how do you negotiate that conversation, John? Cause you seem like you've been up against this for quite some time. I'm so glad you asked that question. So, so I think there's a Mythology that IBAs replace people, and the answer is they don't, right? What IBAs do is they replace mundane jobs that are low value, high volume, and it turns out that those are the kinds of jobs that actually impact that investment that most contact centers have made in their people. Nobody wants to answer the same question day in and day out. It's what I describe with my customers as a soul sucking job. What we've actually seen is that when you implement RBAs and you do it appropriately, and we'll have to touch upon that in just a second, but when you deploy them appropriately, what you see is that. Customer interaction time with a live agent goes up, so it doesn't actually reduce the time, but they go up in terms of that interaction because they're now having more meaningful conversations on when live agents are typically surveyed what they want to do, and everybody wants to do is a job where they make a difference. So being able to produce a relationship, high value, Thank you. You know long term, you know, engagement with an end customer is what live agents are looking for. So their hope is, is that if you do it right, you're going to actually improve the people that these contact center managers are managing and make their jobs more fulfilling by removing all those mundane soul sucking tasks that an IBA is better handled. Yeah, John, that's a super high EQ response because there, there is kind of an assumption that, you know, the only job that we're doing right now is the entire job that needs to be doing. It's a weird statement, but, but if you were to take away those mundane tasks. Could you actually get to the job that you haven't been able to get to, which might be better account based management, better outbound calls to just provide offers and to drive business inside the organization. There's so much that isn't being done because the capacity to do it is not there. And when you. Move the mundane tasks, which are sucking at that capacity, right? And you move that over into another area. You can actually maybe move the ball forward better for the organization. It's a it's a very nuanced topic, and it requires Some kind of big brain thinking, you know, you really have to kind of get out of your own space and think about like, what could we be doing different in the business? I think that's the disruptive nature of AI. Yeah. And we're going to ask our customers to rethink how they do business because most of our customers are used to a paradigm that limited them at the beginning with just IVRs. So let's maybe segue a little bit into another topic, kind of back to AI. Okay. And trying to understand a little bit about how AI and RingCX align, artificial intelligence and self service solutions have gained obviously immense traction. The majority of the mind share and topic discussions are about AI. And. That is no different with inside of the contact center industry. Like I said before, these topics are the top of the marquee and businesses recognize the potential of AI and self service to transform their customer service and empower customers to find answers quickly. But with all of its promise, we could probably agree vastly misunderstood in terms of its application and most providers tout their AI capabilities to boot and you are no different. Can you provide some insights maybe into what AI comes out of the box with RingCX and what possible options one could have in extending the AI capabilities. So yeah, AI, it's, is an extremely broad term because there's a lot of technologies that, that are kind of rolled up under that, that, that label. And there's, there's like three general categories of AI. There's generative AI, which is what chat GPT is kind of. democratize and made everybody aware of even the consumer who, who'd never, you know, looked at the technology before. Then there's conversational AI, which is, you know, where you have a chat bot or a voice bot that allows a customer to basically communicate back and forth with the machine. And it'll try, it'll use natural language, understanding and natural language processing to translate what I'm saying. It is something that the bot can respond to. And those are very useful for things like, like heavily. Repetitive type of interactions that an agent typically deals with. Yeah, certainly. So I think RingCentral probably has an interesting take on this and it really began with redefining the customer journey. I think historically contact centers have thought about the customer journey. The moment a customer picks up the phone and calls us, that's when the clock would start. However, over the last, call it, 24 months we've really moved that timeline back. And we want to start the clock the second the customer has the question pop into their head. Right? It could be, where is my order? It could be, what time is the store open to? You know, who won the ballgame last night? Whatever that case is, right? We want to start solving with that information right then and there. For And that extends all the way to what's happening on the website, what's happening in the Google search. And what we've learned is the better we control those upstream activities, the bigger the impact to the contact center and doing things like reducing call volume, even just having happier customers who call in because they're not frustrated. Those qualitative benefits are, I mean, just incredibly both hard to. To account for, but also the benefit to fixing that exponential. And so we've developed a multi stage solution at Ringcentral that really allows the customer as they go through that journey to keep feeding them additional information to get them to their ultimate goal. And it's been very successful so far. All of these channels were sort of designed with. AI in mind and being able to deploy a bot. So I think that's a nice segue into what we were just talking about is you don't have to be afraid of turning these channels on and then overwhelming your contact center with volume. There's a very smart way to do it while deploying AI to be able to have this be a win win all the way around. Yeah. It's really interesting. As we moved into a digital economy, so much of how people transact and how people communicate is done in a mobile sense. So, you know, if you're going to go to a hotel, you're not waiting to get. To the hotel and check in and have a conversation about something that you might be wanting to do there. You may actually be doing that in the back of the Uber on the way to the hotel. This is kind of the frictionless environment that we're, we're moving into. And if your technology does not meet people on those various different mediums, you're going to lose out and people are going to select other. Companies, other brands. And we know that ultimately today, right now, it takes one bad experience for a person to completely drop, maybe a lifetime of doing business with that company, just because they don't feel that company is in touch anymore, or they don't value their business anymore, or they created potentially a problem in their schedule or in their flow of the day that they can't have that anymore. So they move to. Another provider that will meet them where they need to be. And it's just that tolerance is so little now that I don't understand sometimes why there is a reticence to want to bridge beyond the email bridge, beyond the simple chat. Bridge me on the voice and to get into all these other channels. I mean, right away, you will probably maybe open up a new audience and maybe serve an audience that's underserved. And you had no idea that they were starting to fall away. So it's a, it's a very good philosophical question that you raised, Frank. They're just starting out. They need IVAs that can do ASR and text to speech really well and handle certain modalities. Well, that might be what we call our standard IVA. And that's 80 percent of the call volume, at least today, for example, with this example customer. But this customer wants to incorporate more natural language, right? So we say 20 percent of the calls going forward might be natural language. So rather than have 100 IDAs that do natural language, and now you've got some capabilities that are essentially not a good fit for 80 percent of the calls, We can mix and match, which means as the customer reaches that tipping point where they're doing more and more with natural language, we essentially just swap out that resource from a virtual IVA perspective. And that's, that's that investment protection I was talking about. Yeah, that's fantastic. And I want to go back to, again, the model that you have, because I do think it's brilliant because most people, when they're thinking of the contact center, you know, it's really about how many took us in seats do we have? And so I have 300 people and and maybe as you look at some contraction Maybe you don't have contraction like you mentioned earlier, but maybe you do have some contraction Well, how does that get refilled by having the fact that you purchased 25 or 30? seats of an iva Psychologically, I believe shows where that fill is like, it's very easy to consume because it's part of the context and the construct of how a contact center works today. So it's easy for people to adopt. And when it goes down to the transactional level, I think it makes it harder. So I just think that That is actually a really, really smart way to approach the market. And I think it's a key differentiator. And I think it also maybe makes it easier to buy. Yeah. Philosophically, we're trying to implant the idea in our customers that this is a resource, very similar to human beings. So if you're trying to buy them, you know, minutes at a time at a rate per minute that really negates that approach. The approach is here's a resource they're available to you 24 hours a day. It's seven days a week. They never get sick. They never take vacation. You can very quickly scale when you need to. You don't want to get you know, painted into a corner or find out that you have very little runway as these technologies change. Five9 with our platform approach really allows you to be able to take advantage of the best out there. And we vet them, which also makes sure that at the end of the day, you're not the guinea pig or some AI provider will vet them for you vis a vis the studio platform, which is where IVAs live within five9. Interesting. And it seems to me that the IVA itself. As a component of the five9 offering is something that, that works together with the contact center, the core contact center. But could you use this IVA with another contact center? Oh, absolutely. In fact, that's kind of was our value proposition during COVID. So I, myself implemented a solution for a company they were using in the bias CS 1 K. They were a large mortgage you know clearing house. So essentially they had a set of customers very much like a BPO. And when COVID hit, they had a choice. I'd send 500 employees home with no real ability to support them, or incorporate IVAs with some other born in the cloud UC solutions to actually help facilitate that transfer from in office to work at home. And so we did that, and as a result, they're able to continue operating. But that connection there was not to a born in the cloud solution. It was to a very old and tried and true CS 1 K, which is a big Avaya, a bio platform. Interesting. And, and shockingly, There are many platforms still sitting out there that resemble that description. Did they have a positive overall sentiment during the conversation? Did they have good energy? Did they interrupt the caller? When it comes to the AI pieces about Cognigy and dialogue flow, I would imagine that You know, part of your professional services teams can work with the client. Cognigy and Dialogflow are independent solutions that a customer would, would subscribe to. And those are those are usage based, right? You, you pay based on how often customers actually leverage that. In our platform, you, you, you set up the link and then you basically say on this digital channel, I'm going to check a box that says, I'm going to handle this with a bot as the first point of, of response to the inbound message. Now we, we will. Not necessarily do, you know, build out those intents and the, and the responses and so on, that's where we could leverage partners like yourselves and other consultants to help the customer with that. If they don't have the, the expertise on staff. So if we're looking at the call flow and if the call flow calls for sending a customer down a different path, and that is a digital path. So we're going to do, we're going to try and work with deflection. We now put them into this bot. If the bot experience requires movement back to an agent, we can reroute them back into RingCX to be escalated to that live agent. Yeah. And, and even if you don't use a bot to handle some of these, some of these low value, high volume type of interactions, which you sometimes run into we do use AI categorization and routing. The bot experience requires movement back to an agent. We can reroute them back into. RingCX to be escalated to that live agent. Yeah. And, and even if you don't use a bot to handle some of these, some of these low value, high volume type of interactions, which you sometimes run into we do use AI categorization and routing. So we look at the content of the message and then, and then basically say, okay, well, based on what's written in the message. The digital message, we can now decide how to, how to route it to the right queue. So for example, I can have a single email channel on the platform, one email address, and then based on what the customer writes, I will route it to, let's say billing or customer service or sales based on the, on the question content. So we use AI for that routing, which then again, You know, is one of the biggest frustrations when it comes to a customer reaching out to a business is how do I get to the right person who has the right expertise to handle my, my, my question and, and these days with, with agents being asked to do more with less, many agents are actually. Being tasked to be able to handle more than just one specific type of call. And so how do you empower an agent? One of the things that brings CX brings to the table is we have an embedded knowledge base feature so that I can now have articles written that will help an agent and that they're context sensitive. So we can essentially say, you know, based on, on what channel, what queue the call came in on or the interaction came in on, it'll pop records or in the knowledge base that are relevant to that makes it very easy for the agent. One of the things that we see in conversations around AI in particular is obviously, you know, part of the talk track that goes with it is the displacement of people on the contact center. And that can be reapportionment or reassignment with inside the business, but regardless, it has that connotation and perfectly honest. Many of the contact center leaders that we work with are not too excited about that particular element of it. That it's there. The group of people that they work with has become a family. And. And the thought of downsizing that family as much as it may be for the betterment, sometimes of the overall organization from a cost perspective and a sustainability perspective is a challenge. And obviously AI doesn't have to mean that. How do you approach that? You know, as you're, as you're talking about the impact of the customer experience, that's one thing, but then what are some of the business impacts and, and how are people receptive to some of those various different. Points or optional points that they have in terms of course correction. Yeah. So first I want to be mindful that I'm not underplaying that this is disruptive technology. I think if you look back in history, the printing press, the cotton gin, right. Like all of those things have created seismic shifts in the labor force. And AI I think will eventually do that. I think in the short term, you know, I think everybody doesn't have to be overly concerned that this is just going to wipe out the team immediately, but I don't want to be naive and say that changes aren't coming.

Head Shot

Frank Rogers

Host