The ActivateCX Podcast
Join Frank Rogers on The ActivateCX™ Podcast, your resource for demystifying, clarifying, and providing guidance around AI, CXM, and the modern Cloud Contact Center.
In this Podcast series, Frank interviews Thought Leaders, Unpacks critical AI & CX technology, and addresses the leading Experience topics of the day.
#cx #customerexperience #ai #ex #cxm #contactcenter #salesstrategy
The ActivateCX Podcast
Why Your AI VoiceBots Drive Profits and Customer Satisfaction
Learn more about using AI to improve CX: https://activateCX.arroyo360.com/ai
----
Want to improve customer experience, learn how AI contact center solutions can help! In this video, we interview Michelle Schroeder (at PolyAi) CX Industry AI expert! Together Frank and Michelle dive into the world of conversational AI as a means to boost your sales and customer support performance.
Michelle also addresses prevalent fears and uncertainties surrounding AI, discussing ethical concerns, privacy issues, and the potential for workforce displacement. At the end of the day, learn how to keep your customers and become sustainable against competition?
Don't forget to subscribe for more tech updates!
#podcast #contactcenter #ai #customerservice #salesstrategy #voicebot
Michelle, welcome to the show. Thanks for having me, Frank. It's great to be here. So I just want to say that, I love the fact that, we have the ability to look at industries from different perspectives and yours is unique, you're the marketing and the brand leader for. a CX focused AI technology company and not a small one at that. And I'm just really interested right off the bat, just in terms of how do you spend your time between building brand, but at the same time staying abreast of not only your own product, but where the industry's going. Yeah, for sure. I mean, the honest answer is that those things all need to happen at the same exact time. And all of those things need to inform the strategy. To the same exact degree of weight. And I really try to avoid splitting them up as much as possible. You can't have a marketing vision without a continuous loop that ties from your customer, from the industry and from the reality of like what technology can offer today. And then, as you know, in our space, there's so much noise. There's so much change that you sometimes feel like in that loop, your butt heads down kind of underwater. And you need to kind of come up for air with some very good questions about what's happening. And so my come up for air questions are always things like, is this real? Is this a trend? Do our clients care? Do their customers care? What's the actual impact? And the what's real question, I think, is the same one that a lot of prospects get caught up in. And so we try to focus on answering those questions and then use those answers to infuse into our campaigns, into our content. And then reconcile it with what's actually true and what feels honest for our brand. A lot of times, especially in AI, marketing gets a pretty bad rap. Because a lot of it just doesn't feel authentic. It feels like everybody's kind of jumping on the same buzzwords on the same language to describe real work that AI is doing for sure. But I always say, like, you shouldn't frost a cake that tastes bad if you know it's your, it's a waste of frosting. And so, like, marketing a lot of times should be about validating whether or not that cake is good, whether or not the market actually wants what you are baking whether or not, the technology underlying it and underpinning it is real. And these are things that I believe deeply about at Poly, and it's why I love my role here and why I'm excited to talk about it. There seems to be on one hand, definite excitement, about the possibility of this technology and what it can do. But at the same time, there seems to be a lot of fear, uncertainty and doubt, just a lot of FUD that's out there in terms around ethics, privacy, displacement of people. How do you assuage concerns, , as a marketer? And what do you think are the most credible concerns that people Probably should have like that aren't just things that are just out there in the ether, but like maybe they should really be thinking about these things. Yeah, definitely. I mean, I think we've been hearing for decades from Hollywood, from literature, that AI is going to be either the end of the world or the end of our jobs as we know it and FUD,, like fear, uncertainty, doubt. It's like very beyond natural response to all of this. But what I think is really interesting is that a lot of buyers of conversational AI s. Like really narrowing in on that space don't know exactly what they're afraid of. So many startups, for example, in this space are a screen kind of in front of what people are actually worried about, which is the underlying foundational models. So their ethics, their security, their privacy approach and many of our competitors, like many of those thin wrapper conversational AI startups don't really have a whole lot of tools to battle. That yeah. The things that are kind of underpinning very, very real uncertainty, and that is genuinely scary. We're trained to kind of look at the catastrophic as business owners. It's how we survive. It's how we adapt. It's how we protect our customers. So like an AI privacy breach or a hack, absolutely worthy of our concern. But there's a way more immediate threat, I think that's in front of everyone. And you can probably guess what it is. Can you? A threat to adopting AI, I think would be, over reliance upon the AI, where individuals are not truly vesting into really what it takes, because it's not very well communicated, to be perfectly honest, what it takes to deploy AI. 100%. Yeah. So it's a journey that you're going to start. You're not handing it off to a machine and walking away from it. Absolutely. Absolutely right. And I think that the downstream impact of that is the customer. The person who loses in that equation is the customer. And so right now you have a lot of people flocking toward automation as a way to kind of. Solve some immediate pain in their contact centers. For example, staffing, labor, all of that stuff is really, really difficult. And so you're almost trading this long term fear of something going very, very wrong for the short term pain of not being able to sort of meet that labor demand or meet that customer demand in your contact center. But The real risk that's happening on a day to day basis right now, it's happening for you and I, whenever we call into a business and we're frustrated at the automation loop that you get stuck in, is that that relationship between a business and their buyer is risked by bad AI deployments. And so as Yeah. And I think as much as we are afraid of AI the reasons why we're afraid. I think it's again, just very, very easy to look at the big picture and not just the thing that's right in front of all of us. That bad AI adoptions reduce your loyalty. They reduce your lifetime value. They reduce your credibility in the market. And getting that right means is a very, very, very big and impactful thing. It's interesting as you as you think about it. What it takes to be a good marketer. I think that there are some very similar attributes to being good at deploying AI. That is, you really need to understand the problem. And you need to understand the pain. And then you need to understand what the symptoms of that pain are like you're feeling this and then and then be able to Contextually place what you do within the context of that pain and help solve a problem for somebody I feel ai is very similar. It has a a similar flow to understanding a user story, being able to understand the intent of the call, being able to understand language and what that language means and translate that into a journey. It's a very, it's a very deep process. But when you go through it, you understand your business fundamentally more than you ever did before and your customer more than ever before. Would you agree? Absolutely. I think it's, it's one of the so AI done for automation for customer support is one of the most transformational use cases that exists for an enterprise, partially because of what you just mentioned, to do it really, really well, you need to think about all of these systems. I always say to new clients, like you, you want to think about it the same way. As hiring a human being like hiring a customer experience leader in your business. What type of context would they need to make good decisions to hire to staff, you know, to train? You need to think about the same types of things with AI. And the more of these systems you integrate, the better of a job you do to meet your customers. you know, where they are and with what they need from you, the more you're going to learn directly in the voice of the customer. So it doesn't just become this amazing kind of command center for you, where you can deploy campaigns, where you can query problems, you can better understand your customers, but it also becomes like a. Insights database for you to learn about what types of changes you should make as a business. How do you reduce calls to your call center? how do you think about improving the things that are making people call you in the first place? And this is one of the biggest kind of advantages of having digitizing voice and bringing it to the same level as your website, for example, or an app. This is one of the biggest advantages, but it takes more than just the contact center leader to drive that transformation. It's an extremely cross functional. You know, process. I couldn't agree with that more. Typically, it's funny how technologies are purchased, right? Like maybe the CMO is purchasing or maybe the VP of customer experience or customer support more likely. And that may only represent one aspect of the customer journey. And I would agree that this is a time when we're looking across technology sets and AI. I believe it is such a cross functional problem to solve. And that is really kind of the key is as AI Consulting professionals, so to speak, or people that are in in that particular industry need to be particularly good at in terms of rallying the troops and moving towards the understanding of a use case and then fundamentally how you're going to approach that together. So I think that that and getting maybe a set of like shared expectations of how this is going to work. Yes, for sure. So if we look at organizations adopting AI. And again, we're talking about conversational AI. So chat bots, voice bots, if we're going to bring those into an environment that probably is already very human skewed, right? We're going to end up with a hybrid experience at some particular point. What's the tipping point in that? What is too far? What is too little? How do we negotiate that? Yeah. I mean, I think the only answer to that question is that it's going to depend on the enterprise. There are going to be some very high value, high touch clienteling type businesses, like think about a luxury jeweler where not having a human as part of that experience probably won't work very well. For your, not just for your clientele, but for your brand, for who you are, for your ethos as a company that said, you know, I do believe that we will always need people as part of your support strategy at the same time that I believe that it's possible to automate a hundred percent of it today. I think it's, it's possible to automate a hundred percent of your support inquiries fairly well today, but that road. to getting to 100%. It's going to be very enterprise dependent, but it's also going to be paved by people by smart people who care deeply about their customers. And then just sticking with the analogy of like pavement. We all know, like once a road is put down, it's going to need repavement and improvements, and those generally are done by people. So even for a customer roadmap, you think about the journey and how it might change or the brand and how it might change. A person does still need to think not just one, but Possibly tons of people will need to be transitioned from that agent role like picking up phones or that team lead role or that leadership role and transitioned into kind of an A. I. Management role. Somebody today in our systems with our deployments are listening to calls and flagging things within our model flagging either. Hey, I didn't like how you said yeah, and not yes, that doesn't go with our brand. And it's not something we want to do. talked about in design right up front, because we didn't think of it. But somebody does need to be there. And that's a very, very tiny example, obviously, but there are things that are that nuanced and things that are bigger that people need to have a part in designing. This is a partnership between people and technology. And in order to get it Right. In order to get that balance, right. It is going to be up to the business. It is going to be up to their goals. But we have customers today that are extremely ambitious and they want to automate absolutely everything, including campaigns and cross promotions of products. And we do that with them, with the people at those businesses and very close and tight alignment. We use their best reps to help train our models, to be better tuned on their unique business, the changes that happen from. you know, one season to the next. We find together basically these best strategies and approaches. And then also, of course, learn from the customers, the other person at the end of this equation for what they actually like, what they respond to, what drives more engagement and what lowers it. And we are constantly tuning and fixing and repaving and improving the road to success. So having you on record is saying that AI is creating jobs. Is that did I hear you correctly there, Michelle? It's improving jobs. In my opinion, it's transitioning jobs. Like I, I ran a contact center over a decade ago, and I remember saying this when I was interviewing here at this business, you know, I'm obviously a marketing and experienced person at heart, so I went in with extremely high aspects. I was like, we are going to turn this into the Ritz Carlton, it was a web hosting company. We're going to turn this into the Ritz Carlton of web hosting. Everyone was like, you have a holiday in budget. And that's, that's good. I'm not putting down the holiday in, but they're just like, you know, this, this is what you have. You know, you need to figure out your KPIs and like making sure that we're hitting our contact center metrics with the reality of why people are calling and how long it takes to solve a problem. Aspirational energy is great. And I think if I had a technology like this, I could have transitioned it into AI. I could have transitioned it into a system that scaled and allowed me to build the thing that I really wanted to build. Without the limitations, I think that I, I had certainly a decade ago. So yes, it's creating jobs, but it's transitioning jobs. I think it's hopefully improving them and turning folks that used to answer calls into AI. You know, consultants, Building on everything you just said there , that we do working with clients and as part of education is walking through understanding the concept of highest and best use, and that's been used in real estate for years and years. But we can apply it towards people inside of an organization. What is the highest and best use of these people inside of the organization, assuming that they have this capacity? This has led to computing labor arbitrage and understanding who's going to be a high value agent and then what people maybe move to another part of the organization where quite literally there is not good supply where there's a difficulty in finding skilled people for those positions in sales, in business development, in professional services. These people working inside of the customer service side of the business already have an understanding of the product, the service, the organization. It's a natural fit. So we're getting into a discussion of how do we manage this new world opportunity organization. So we don't have to look at it purely on the surface of displacement, although some healthy displacement may occur. You mentioned so many things in terms of just making things better for the agents and everything else, that answer that you had just had so much richness to it. If you look at how AI is improving. Efficiency. That's great. But there are people out there that are thinking that there is a sacrifice of empathy, right? In customer service and that that empathy doesn't necessarily support a great relationship with the customer. And one of the things we were talking about in this a little bit around the customer is that consumer behavior, customer behavior has changed dramatically since the dreaded COVIDS. And so now people have one bad brand experience and they're gone. They may they leave without, protesting too much. They leave without a signal. And sometimes they're gone for as long as a couple of years. What do you think about that whole empathy component? Yeah, I mean, I think about empathy a lot, especially in the context of a I may be maybe too much. I was speaking with one of our dialogue designers on the team about this a couple weeks ago, and we were saying that, like one of the Distinguishing features of empathy like human empathy is that sense of I know how you feel, you know, I get it I know how you feel like I might not have had the exact same experience, but I but I understand AI Cannot know how you feel But what makes it empathetic is that the people who designed it do? They do know You know what it's like to kind of be in that position and they've put in that effort they've designed conversations that feel immensely human because they are immensely human that that's who's behind You know that technology and she was saying something really clever about you know, can we put easter eggs into our? Experiences that just remind people that behind this automation that sounds so incredibly lifelike which is by design. But there's a person behind it. Like, can we do things that sort of, like, signal to that, you know, caller that there's, there's a person behind this design that does know how you feel and designed this on purpose. That said, I'm thinking about this and going back to the fact that, you know, I led a call center a long time ago. There were scripts for empathy, especially for very, very difficult callers. We literally scripted every word. So if you knew that. on the other side of that call, that this person was reading from a script, well, hopefully, naturally, hopefully would that, you know, take down your sense that this person understood how you felt? And I think, You know, I think it's probably performance based, but but I think not designing with empathy is a lot about the delivery. It's about meeting the customer where they are. It's about not making them wait. It's about, you know, like recognizing and understanding the context that they're bringing with them when they bring you a problem. I've got a great example of this that I'd love for you to play. I sent you this call a little bit earlier. I, I do have it right here, I'll just queue it up, we'll just let it run. Phone ringing Hi, thanks for calling NSNI, how can I help? Oh, hi! My mom's been divided up, it's the probate never will. I'm very sorry to hear of your loss, our dedicated team is here to help. Just so I'm sure, have you already sent your instructions to us? Sorry? Have you sent us a Bereavement Claims Form? No, where is that? Okay, all you need to do is get the executor, legal representative or next of kin to fill out our online Bereavement Claims Form. Do you know where to find this? No. No problem, you can find all our forms on our website. Are you able to go online? Yes. Great. I'll send you a link to all our forms. Am I okay to send this to the number you're calling from? Yes, that'll be fine. Okay, I've sent you a text with this. Is there anything else I can help you with today? No thank you, that's been fine. Thanks very much. Thank you for calling NSNI. Goodbye. Bye. BEEP BEEP So You can hear, you know, how much empathy was just in the, the voice and in the delivery and in the design, but also in the fact that she got what she needed. And when she didn't understand the first time when he said, you know, have you sent your instructions in, you know, he rerouted it. He rephrased it in a way that she understood. He met her where she was, he sent her those forms and made it a little bit easier. in a very difficult situation to get to the next step so that, you know, she could move on. And that's empathy by design. And that design is by a person with human empathy. And I think that that's something that's easy to forget in the AI context. And you do have to remind people with a very human voice and tonal changes and, you know, rephrasing and all of those types of things. But that's the That's the value of having a conversational AI partner with proprietary tech. That isn't just a wrapper kind of around chat GPT. Right. Yeah. I would say that, once you get your, empathy component of your conversational AI voice bot dialed in, that it may be easier than training people with little empathy to have empathy. Yes, I agree. There's a humorous dichotomy of making the, the compute more human. You know, on the flip side of that human empathy. can also be seen as a weakness. I don't want to call out the brand where this happened, but there was a very, very, very large data breach that happened because somebody played on the heartstrings of a human agent and said, you know, X, Y, Z happened to me, like my life's destroyed. I really, really, really need you to help me reset my password. And. The woman heard the sob story and she felt really badly for this individual who probably acted it super well and, you know, changed his password and this person got access to millions and millions of customer data instances that should have never happened. But I mean, that's, that's not a story you really want to align usually with empathy. Empathy is not a weakness, but for a brand with sensitive data. So what are your thoughts about the generations sitting out there, there's data that says Gen Z won't even deal with a company if they don't have a self service option. They're like, I don't want to talk to anybody at all. But how much of that is pervasive across millennials and boomers. They're all on the internet, they're utilizing e commerce, social, digital experiences. They're getting used to this kind of level of access and almost a level of triage with respect to reaching out. Where do you see that in terms of Generational.? Yeah. I mean, I think self service is step one for a lot of us, the cross generation generationally. If you can't find the answer on the website too, on their self service, you know, platform, you might go, for example, to Reddit or any sort of social media to see if you can't find your answer. Generally, when people are calling it's because they weren't able to find the answer themselves. It's a higher order problem. And the whole goal, I think of, of voice and. You know, like doing this well, doing automation well is meeting the customer where they are putting the voice channel on par with a digital one, maybe making the voice channel a self service channel. It is like our CEO says this all the time. It's the most intuitive channel. And we started to study this on our side to Gen Z like millennials. Will pretty much always try to find the answer on their own before hopping to a call for like a first order, kind of lower level problem. But if you offer them chat while they might use chat, it actually takes longer to translate your problem from your head to your fingers and type out, you know, your question and you tend to put less information in a chat in your first query in a chat than you would if you just like right now, I could probably answer this a lot shorter if I had pre written it. But here we are, here I am rambling. But you get a lot more richness and contextual data from the voice and Gen Z. We just did this study a few weeks ago and Forbes picked it up which is that Gen Z does actually call in as much, if not more than millennials, because generally speaking, they tried to find the answer on their own and now they're calling in to support. And so our job is to make that call into support as good as if they still had it. Got the answer themselves. So, for example, near instant response from an agent. No, no waiting, like no hold times at all. Real time customer context that pulls in your history and maybe even preempts why you're calling in the first place. We do this for some of our clients. It is It's such a hit with customers. If you call in and you say, Hey, we actually see the shipment is delayed. Is this what you're calling about? Yep. Well, can I text you the details? I'm like, when you're going to get it, please, you know, and that's, and that's the entire call. It feels like self service. It feels as good as if you found the answer yourself. And then this, this third piece, just a high degree of accuracy in that call. Partially why you call in to speak to somebody because that someone is going to care a little bit more deeply about your problem and try to get you to an answer and you know that their metrics, their KPIs are tied to answering the question for you. If we can bring voice to that same level as if you were trying to find the answer for yourself. I mean, I think cross generation. I can't seem to get that word right. Cross generationally, you will see success no matter what your age is, no matter what your preferences are. I, agree with that. It's very impressive to just taking that approach of being proactive. So now I'm saying, and how can I help you is just the fact that you've got enough data on that individual as to what's going on with their experience right now to probably have a pretty good idea of why they're calling. So. I think that's great. So if we look at, you know, maybe some compelling stories you might be aware of across where a bot in your opinion, and then it seems strange to say above and beyond what you would expect, in a customer experience, and maybe it's customer support, but maybe even something that has to do with revenue generation as well. So this one's going to sound a little bit paradoxical just because the brand lost money on this call. But when I think about like great support and like the best support experiences I ever had, it was because the person on the other end. made a strategic concession for me. They did something for me and it might have cost them something, but they made that choice to do it either because I was high value or I was upset or I was really nice to them. And so they just wanted to do something nice back, but they made some kind of concession for me. And we did this for a company who on their website had a price matching You know, policy, they would do price matching if you brought it into the store and you brought your seat and you told them you found the price lower elsewhere. And somebody called in and said, Hey, you know, you've got price matching on your, your site. And we'd like to just, we'd like to call and basically get our, our, our product adjusted. And we did it for them. And then it turned into another use case for us, which is, you know, just, just on adjusting prices for people that were calling in again, like while the brand. is losing money on those types of calls. They did the math. They know that that's an ROI positive loss because it ups loyalty and lifetime value. That person keeps coming back. They trust this brand more than any other brand. Instead of shopping elsewhere, they would choose to continue shopping there, even though it's more expensive, even though it's one extra step of calling in on that price adjustment. And then on the back side of that, on the back end of that, They learned about where price sensitivity was in their merchandise. So where are people looking and shopping for different prices and calling in that can help them with price sensitivity that can help them with adjusting their prices and making sure that they're lower in some spots where they're getting a lot of those types of calls. Those types of things, those types of concessions, they make customers really, really happy, but they're also incredibly strategic for the business. Yeah, that's powerful. I mean, the, that kind of almost was leading into my next question. You, you kind of, I don't know if you were telepathically kind of reading my mind here. How do you ensure that ultimately your AI is been configured to be customer centric. And there, you were just describing something that was customer centric. to a fault. But ultimately it was the value proposition and the assertions that you've made as a company, as a marketer, you're actually fulfilling on that promise, which is typically where it falls down. Like there's some sort of marketing, whether it be an ad or something on a website purports to be X. And then somebody gets into a call with somebody and no, it turns out to not really be the case. And so what you did there, like Obviously ended up being something that probably got refined over time, but it fulfilled on the value proposition and the brand. And so the brand had true credibility in the marketplace. that's very powerful. There's this thought process that we will lose relationship with our customers by using AI. And I don't believe that that's the case. Most customers really want to get a transaction done or they want to get a problem solved and it may be that they like talking to Joe or Betty at that organization, but above that, they appreciate more the fact that it gets done. Yes, definitely, definitely. And unfortunately, like we've looked at the contact center as this problem processor. It's a place where customer problems go and they get sort of solutions. They, you know, they go through a workflow and that workflow might include people. It might include automation, might include self service chat, whatever. But we should be looking at it as kind of like a real time monitoring system between brands and customers. And I think to do that, well, you need brand and CX and marketing. You need those voices. in the room while you're thinking about designing this. And that is very rare. I can say as a person who, you know, knows our competitors fairly well and how they sell, who has watched, you know, us do some very large enterprise deployments over our lifetime, really our buyer. Tends to either be somebody who's technical sometimes, you know, in the C. I. O. Space that sits between business and tech or the contact center alone. But really, my hope for this industry is that it starts to really spread its wings and touch a lot more of these teams that care deeply about the customer just from different angles. There's no reason that a contact center cannot transform into something that's helpful to all of those teams that can be more than a problem processor that can drive revenue. not just in these revenue protective ways, like, Oh, well, we answered this call. And so we didn't lose the customer. The CSAT didn't drop. And so we feel safe, but instead anticipating and being productive for that business and, and thinking about all of the ways that you can make strategic concessions at scale or find these opportunities at scale or teach multiple teams about, you know, different process weaknesses or opportunities that tend to exist across a business and give them a means of measuring. Whether or not they're making progress toward fixing those, those types of things, like your contact centers, the richest source of data you have on your customers, because it's all in their own words. And it's all right. Usually in response to issues that they're having with your, your business. And it's typically very siloed and to your point, and this is actually something that we do at Arroyo 360. We have a framework called activate CX, and it's really all about getting from point A to point B. But with the AI that we work clients in terms of formulating user stories, you're right, typically they all come from one or two individuals in the business that have a specific need, but if you look at working across the customer journey it pulls in the CMo.. It pulls in the CRo. It pulls in all these other IT different people that are stakeholders in this process because it's part of the journey that you want to be consistent. and congruent and support that value proposition because I would have to say that probably one of the biggest disconnects amongst many in any type of business is that what's espoused by marketing gets augmented, modified, changed with the sales team does not make it to the same value proposition and values that marketing wants to have in place. Inside of the delivery team that then leads for adoption, then leads into ultimately the advocacy side of the business, which may be so far downstream, they never got that memo. And, and ultimately at least to like, A break in relationship because you set, you set an impression and an expectation and then you don't deliver on it. It's a painful situation. So let's shift gears. Let's talk about like the mouse in the house right now, which is agentic AI that's Oh, yeah. the next kind of thing that's going to happen. It's happening now, but it's also a journey itself, right? It's going to evolve. What's got you excited about that? For sure. I mean, I'd be curious to get your thoughts on this as well. So maybe we can, I'll flip it after. I think certain tasks that generative is not as well suited for will get better as a result of agentic AI. I think, you know, for example, API calls to systems that are integrated with. You know, with, with us to be able to serve customers better by pulling context from multiple different sources, that whole landscape will get flattened by agentic, you might have, you know, a partner offering an agent to do the types of things that you used to have to make calls for, and then make rules for how, you know, you. do not pass go unless you have actually done this or completed this task or received this bit of information or token from, you know, whatever, whatever system you're integrated with. And I think agentics already starting to show us some improvements there, but I think one of the most interesting applications here is around customer data and customer context. The same way we have an agent that serves the customer, you know, that that's on the phone and, you know, talking with them or chatting with them and solving their problems, you could have one that's understanding and picking up insights and thinking about and reasoning about everything the customer is telling you and all the first party information that you have. And so whether or not that, you know, customer agent is in service of the You know, agent that's speaking to the customer that's engaging with the customer, or if it's in service of the business, and, you know, intelligent, much more intelligent dashboards than we have today, and actually pulling interesting information and making suggestions. I don't know. But this is something that we are very, very interested in right now. And we are piloting in very, very interesting ways, which I will leave for another discussion that said, like, how are you seeing agentic AI and are your, are your customers asking you about it? No, I don't think it's reached, you know, everybody's mailbox yet, but I, I'm very enthusiastic about its application for, for many reasons, because, you know, so much of generative AI right now has been about taking care of that spot issue. Done. And then assessing. That spot issue did, you know, in a call summary or in some sort of sentiment or post survey. Okay. How did that work? There is a broader context to the ramifications of anything that we deal with a case case in point might be a person calls in, maybe they talk to a generative AI, they're in a chat bot or they're in a voice bot and they're having a bad day. They're mad and they're not going to be happy, even if they get things solved, they're still ticked at maybe even having to deal with this particular situation, maybe rightfully so, but ultimately they come out of the experience not feeling good about it. So sometimes that can just kind of sit. And sit there for analytics and maybe conversational intelligence. Fundamentally analytics application side of this picks it up. There's, there's no direct action that comes out of it. So I think that the automation and the workflow potential aspects of it would be great. Cause right now we use a CDP to kind of fundamentally. Take this data and activate some sort of action, but the A. I. Could be much more proactive and prescriptive and saying, wow, this was a major problem with this customer. We resolved it. Maybe that was an outcome, but they're still unhappy. We are not going, we're going to remove them from a sequence in our marketing automation platform that just says happy, happy, happy, where they all look at that and go like, I'm not so happy. I'm now even mad at you for not knowing that I'm not happy and sending me this garbage. And maybe we update something in a CRM where maybe there's an account manager or something like that can look at and say, like, I'm going to personally call. Mary and say, man, we messed up and your business is important to us. This is what we want to do for you. And so I think that agentic AI has so many user stories and use cases that are to be discovered. I think that ultimately as human beings, we create an innovation and then we think we know what it can be used for. And then we're surprised that it can do so much more. Yes. I had a funny conversation about why. So going back to the very first question you asked me where we talked about like how you reconcile all of the different, you know, sources of information that could feed into a marketing campaign. And I remember agentic as it started to blow up with agent force and, you know, all of the subsequent kind of blog posts and podcasts and company, like, you know, product pages that were changing and now saying agentic we had a conversation about why. And whether or not we believe this was trend or a real thing. And I think that agentics, one of those things that gets defined differently no matter who you, you talk to about it, but the reason why it's so resonant, in my opinion, is because it approximates the human aspect of getting jobs done and agent. For a task, it's a very, very palatable and like easy to understand framework for this other thing, which is like conversational AI, it's all of it, you know, it's, it's data, it's insights, it's automation, it's, you know, transactions, it's API calls, it's all of this stuff all at once. And it feels like you're buying a big giant black box and agenda does break it down for you. It does. And I think it's like with all things in AI. When you have nuance and context, those two things, which are human qualities they essentially are the result of discernment. And when that can be coded and codified that's a pretty miraculous process. So when you look at the future landscape of customer engagement and AI, the Continuous rise and use of it. What do you think businesses should be preparing for? What should their mindset be? If you were just to give them a little bit of coaching advice, sitting on the bench with them and saying, it's going to be okay, this is the way I think you should think about this"! What would you say to them? Yeah, I mean, I would say, you know, do not focus on short term pain when you're designing something for the long term, maybe even go a bit further, like go to step two, step two of automation isn't just, you know, solving for containment or making sure that, you know. A customer gets their issue solved. It's everything we've been talking about this whole time. It's cross functionally engaging with your peers to find out how you can make automation work for you and work for them in a way that's much wider, has a much bigger impact. Designing for your customer means that you actually have to take their words, their insights, their feedback, you know, into account when you're designing. And I think if you. Design with the end state in mind, you think about automation and conversational AI as you know, what we've been talking about this this whole time, which is building it as a command center for you, not a problem processor, but something that can surface for you, all of the things that you can do to improve your relationship with your customers, your processes as a business, what might that look like? You know, what types of dashboards or information would you want from your contact center? What questions would each of these cross functional peers that you have, what would they want to know from the contact center? And then build something that can help you answer that, that can help automate enough volume for you. And certainly delight enough of your customers that you've earned the right to have that engagement that you actually get enough information out of it to do something really interesting with your business. Because this first level automation, this like quick well, we're going to do intelligent routing, or we're going to automate FAQs. And if you don't get that perfectly, right, you wind up. damaging your customer relationships and for what, for how many cents hourly of savings over having a human, you know, filled contact center. I do think that we all have responsibility in making sure that this next generation of automation is better than it's been. And a lot of that is shared, not just between, you know, the vendor, but the enterprise like let's aim higher. I think one of the best signs of maturity I ever see is when a question is answered with another question. Michelle, thanks for being on the show. Thanks for having me, Frank. This is a blast.
